Abstract
Since the increasing popularity of large language model (LLM) backend systems, it is common and necessary to deploy stable serverless serving of LLM on multi-GPU clusters with autoscaling. However, there exist challenges because the diversity and co-location of applications in multi-GPU clusters will lead to low service quality and GPU utilization. To address them, we build ENOVA, a deployment, monitoring and autoscaling service towards serverless LLM serving. ENOVA deconstructs the execution process of LLM service comprehensively, based on which ENOVA designs a configuration recommendation module for automatic deployment on any GPU clusters and a performance detection module for autoscaling. On top of them, ENOVA implements a deployment execution engine for multi-GPU cluster scheduling. The experiment results show that ENOVA significantly outperforms other state-of-the-art methods and is suitable for wide deployment in large online systems.
Title:
Exploring Thermography Technology: A Comprehensive Facial Dataset for Face Detection, Recognition, and Emotion
Authors: Mohamed Fawzi Abdelshafie Abuhussein, Ashraf Darwish, Aboul Ella Hassanien
Abstract
This dataset includes 6823 thermal images captured using a UNI-T UTi165A camera for face detection, recognition, and emotion analysis. It consists of 2485 facial recognition images depicting emotions (happy, sad, angry, natural, surprised), 2054 images for face recognition, and 2284 images for face detection. The dataset covers various conditions, color palettes, shooting angles, and zoom levels, with a temperature range of -10°C to 400°C and a resolution of 19,200 pixels. It serves as a valuable resource for advancing thermal imaging technology, aiding in algorithm development, and benchmarking for facial recognition across different palettes. Additionally, it contributes to facial motion recognition, fostering interdisciplinary collaboration in computer vision, psychology, and neuroscience. The dataset promotes transparency in thermal face detection and recognition research, with applications in security, healthcare, and human-computer interaction.
Title:
AI-Based Copyright Detection Of An Image In a Video Using Degree Of Similarity And Image Hashing
Abstract
The expanse of information available over the internet makes it difficult to identify whether a specific work is a replica or a duplication of a protected work, especially if we talk about visual representations. Strategies are planned to identify the utilization of the copyrighted image in a report. Still, we want to resolve the issue of involving a copyrighted image in a video and a calculation that could recognize the degree of similarity of the copyrighted picture utilized in the video, even for the pieces of the video that are not featured a lot and in the end perform characterization errands on those edges. Machine learning (ML) and artificial intelligence (AI) are vital to address this problem. Numerous associations have been creating different calculations to screen the identification of copyrighted work. This work means concentrating on those calculations, recognizing designs inside the information, and fabricating a more reasonable model for copyrighted image classification and detection. We have used different algorithms like- Image Processing, Convolutional Neural Networks (CNN), Image hashing, etc. Keywords- Copyright, Artificial Intelligence(AI), Copyrighted Image, Convolutional Neural Network(CNN), Image processing, Degree of similarity, Image Hashing.
Title:
Aligning Models with Their Realization through Model-based Systems Engineering
Authors: Lovis Justin Immanuel Zenz, Erik Heiland, Peter Hillmann, Andreas Karcher
Subjects: Subjects:
Systems and Control (eess.SY); Multiagent Systems (cs.MA); Software Engineering (cs.SE)
Abstract
In this paper, we propose a method for aligning models with their realization through the application of model-based systems engineering. Our approach is divided into three steps. (1) Firstly, we leverage domain expertise and the Unified Architecture Framework to establish a reference model that fundamentally describes some domain. (2) Subsequently, we instantiate the reference model as specific models tailored to different scenarios within the domain. (3) Finally, we incorporate corresponding run logic directly into both the reference model and the specific models. In total, we thus provide a practical means to ensure that every implementation result is justified by business demand. We demonstrate our approach using the example of maritime object detection as a specific application (specific model / implementation element) of automatic target recognition as a service reoccurring in various forms (reference model element). Our approach facilitates a more seamless integration of models and implementation, fostering enhanced Business-IT alignment.
Title:
Optimization of Autonomous Driving Image Detection Based on RFAConv and Triplet Attention
Abstract
YOLOv8 plays a crucial role in the realm of autonomous driving, owing to its high-speed target detection, precise identification and positioning, and versatile compatibility across multiple platforms. By processing video streams or images in real-time, YOLOv8 rapidly and accurately identifies obstacles such as vehicles and pedestrians on roadways, offering essential visual data for autonomous driving systems. Moreover, YOLOv8 supports various tasks including instance segmentation, image classification, and attitude estimation, thereby providing comprehensive visual perception for autonomous driving, ultimately enhancing driving safety and efficiency. Recognizing the significance of object detection in autonomous driving scenarios and the challenges faced by existing methods, this paper proposes a holistic approach to enhance the YOLOv8 model. The study introduces two pivotal modifications: the C2f_RFAConv module and the Triplet Attention mechanism. Firstly, the proposed modifications are elaborated upon in the methodological section. The C2f_RFAConv module replaces the original module to enhance feature extraction efficiency, while the Triplet Attention mechanism enhances feature focus. Subsequently, the experimental procedure delineates the training and evaluation process, encompassing training the original YOLOv8, integrating modified modules, and assessing performance improvements using metrics and PR curves. The results demonstrate the efficacy of the modifications, with the improved YOLOv8 model exhibiting significant performance enhancements, including increased MAP values and improvements in PR curves. Lastly, the analysis section elucidates the results and attributes the performance improvements to the introduced modules. C2f_RFAConv enhances feature extraction efficiency, while Triplet Attention improves feature focus for enhanced target detection.
Title:
Unsupervised Anomaly Detection Using Diffusion Trend Analysis
Authors: Eunwoo Kim, Un Yang, Cheol Lae Roh, Stefano Ermon
Abstract
Conventional anomaly detection techniques based on reconstruction via denoising diffusion model are widely used due to their ability to identify anomaly locations and shapes with high performance. However, there is a limitation in determining appropriate noise parameters that can degrade anomalies while preserving normal characteristics. Also, due to the volatility of the diffusion model, normal regions can fluctuate considerably during reconstruction, resulting in false detection. In this paper, we propose a method to detect anomalies by analysis of reconstruction trend depending on the degree of degradation, effectively solving the both problems of existing methods. The proposed method is validated on an open dataset for industrial anomaly detection, improving the performance of existing methods on a number of evaluation criteria. With the ease of combination with existing anomaly detection methods, it provides a tradeoff between computational cost and performance, allowing it high application potential in manufacturing industry.
Title:
BoBa: Boosting Backdoor Detection through Data Distribution Inference in Federated Learning
Authors: Ning Wang, Shanghao Shi, Yang Xiao, Yimin Chen, Y. Thomas Hou, Wenjing Lou
Subjects: Subjects:
Machine Learning (cs.LG); Cryptography and Security (cs.CR)
Abstract
Federated learning, while being a promising approach for collaborative model training, is susceptible to poisoning attacks due to its decentralized nature. Backdoor attacks, in particular, have shown remarkable stealthiness, as they selectively compromise predictions for inputs containing triggers. Previous endeavors to detect and mitigate such attacks are based on the Independent and Identically Distributed (IID) data assumption where benign model updates exhibit high-level similarity in multiple feature spaces due to IID data. Thus, outliers are detected as backdoor attacks. Nevertheless, non-IID data presents substantial challenges in backdoor attack detection, as the data variety introduces variance among benign models, making outlier detection-based mechanisms less effective. We propose a novel distribution-aware anomaly detection mechanism, BoBa, to address this problem. In order to differentiate outliers arising from data variety versus backdoor attack, we propose to break down the problem into two steps: clustering clients utilizing their data distribution followed by a voting-based detection. Based on the intuition that clustering and subsequent backdoor detection can drastically benefit from knowing client data distributions, we propose a novel data distribution inference mechanism. To improve detection robustness, we introduce an overlapping clustering method, where each client is associated with multiple clusters, ensuring that the trustworthiness of a model update is assessed collectively by multiple clusters rather than a single cluster. Through extensive evaluations, we demonstrate that BoBa can reduce the attack success rate to lower than 0.001 while maintaining high main task accuracy across various attack strategies and experimental settings.
Title:
RIO-CPD: A Riemannian Geometric Method for Correlation-aware Online Change Point Detection
Abstract
The objective of change point detection is to identify abrupt changes at potentially multiple points within a data sequence. This task is particularly challenging in the online setting where various types of changes can occur, including shifts in both the marginal and joint distributions of the data. This paper tackles these challenges by sequentially tracking correlation matrices on the Riemannian geometry, where the geodesic distances accurately capture the development of correlations. We propose Rio-CPD, a non-parametric correlation-aware online change point detection framework that combines the Riemannian geometry of the manifold of symmetric positive definite matrices and the cumulative sum statistic (CUSUM) for detecting change points. Rio-CPD enhances CUSUM by computing the geodesic distance from present observations to the Fréchet mean of previous observations. With careful choice of metrics equipped to the Riemannian geometry, Rio-CPD is simple and computationally efficient. Experimental results on both synthetic and real-world datasets demonstrate that Rio-CPD outperforms existing methods in detection accuracy and efficiency.
Title:
Memory Lower Bounds and Impossibility Results for Anonymous Dynamic Broadcast
Authors: Garrett Parzych, Joshua J. Daymude
Subjects: Subjects:
Distributed, Parallel, and Cluster Computing (cs.DC)
Abstract
Broadcast is a ubiquitous distributed computing problem that underpins many other system tasks. In static, connected networks, it was recently shown that broadcast is solvable without any node memory and only constant-size messages in worst-case asymptotically optimal time (Hussak and Trehan, PODC'19/STACS'20/DC'23). In the dynamic setting of adversarial topology changes, however, existing algorithms rely on identifiers, port labels, or polynomial memory to solve broadcast and compute functions over node inputs. We investigate space-efficient, terminating broadcast algorithms for anonymous, synchronous, 1-interval connected dynamic networks and introduce the first memory lower bounds in this setting. Specifically, we prove that broadcast with termination detection is impossible for idle-start algorithms (where only the broadcaster can initially send messages) and otherwise requires $\Omega(\log n)$ memory per node, where $n$ is the number of nodes in the network. Even if the termination condition is relaxed to stabilizing termination (eventually no additional messages are sent), we show that any idle-start algorithm must use $\omega(1)$ memory per node, separating the static and dynamic settings for anonymous broadcast. This lower bound is not far from optimal, as we present an algorithm that solves broadcast with stabilizing termination using $\mathcal{O}(\log n)$ memory per node in worst-case asymptotically optimal time. In sum, these results reveal the necessity of non-constant memory for nontrivial terminating computation in anonymous dynamic networks.
Title:
Semi-supervised 3D Object Detection with PatchTeacher and PillarMix
Abstract
Semi-supervised learning aims to leverage numerous unlabeled data to improve the model performance. Current semi-supervised 3D object detection methods typically use a teacher to generate pseudo labels for a student, and the quality of the pseudo labels is essential for the final performance. In this paper, we propose PatchTeacher, which focuses on partial scene 3D object detection to provide high-quality pseudo labels for the student. Specifically, we divide a complete scene into a series of patches and feed them to our PatchTeacher sequentially. PatchTeacher leverages the low memory consumption advantage of partial scene detection to process point clouds with a high-resolution voxelization, which can minimize the information loss of quantization and extract more fine-grained features. However, it is non-trivial to train a detector on fractions of the scene. Therefore, we introduce three key techniques, i.e., Patch Normalizer, Quadrant Align, and Fovea Selection, to improve the performance of PatchTeacher. Moreover, we devise PillarMix, a strong data augmentation strategy that mixes truncated pillars from different LiDAR scans to generate diverse training samples and thus help the model learn more general representation. Extensive experiments conducted on Waymo and ONCE datasets verify the effectiveness and superiority of our method and we achieve new state-of-the-art results, surpassing existing methods by a large margin. Codes are available at this https URL.
Title:
AraFinNLP 2024: The First Arabic Financial NLP Shared Task
Authors: Sanad Malaysha, Mo El-Haj, Saad Ezzini, Mohammed Khalilia, Mustafa Jarrar, Sultan Almujaiwel, Ismail Berrada, Houda Bouamor
Subjects: Subjects:
Computation and Language (cs.CL)
Abstract
The expanding financial markets of the Arab world require sophisticated Arabic NLP tools. To address this need within the banking domain, the Arabic Financial NLP (AraFinNLP) shared task proposes two subtasks: (i) Multi-dialect Intent Detection and (ii) Cross-dialect Translation and Intent Preservation. This shared task uses the updated ArBanking77 dataset, which includes about 39k parallel queries in MSA and four dialects. Each query is labeled with one or more of a common 77 intents in the banking domain. These resources aim to foster the development of robust financial Arabic NLP, particularly in the areas of machine translation and banking chat-bots. A total of 45 unique teams registered for this shared task, with 11 of them actively participated in the test phase. Specifically, 11 teams participated in Subtask 1, while only 1 team participated in Subtask 2. The winning team of Subtask 1 achieved F1 score of 0.8773, and the only team submitted in Subtask 2 achieved a 1.667 BLEU score.
Title:
IFTR: An Instance-Level Fusion Transformer for Visual Collaborative Perception
Authors: Shaohong Wang, Lu Bin, Xinyu Xiao, Zhiyu Xiang, Hangguan Shan, Eryun Liu
Subjects: Subjects:
Computer Vision and Pattern Recognition (cs.CV)
Abstract
Multi-agent collaborative perception has emerged as a widely recognized technology in the field of autonomous driving in recent years. However, current collaborative perception predominantly relies on LiDAR point clouds, with significantly less attention given to methods using camera images. This severely impedes the development of budget-constrained collaborative systems and the exploitation of the advantages offered by the camera modality. This work proposes an instance-level fusion transformer for visual collaborative perception (IFTR), which enhances the detection performance of camera-only collaborative perception systems through the communication and sharing of visual features. To capture the visual information from multiple agents, we design an instance feature aggregation that interacts with the visual features of individual agents using predefined grid-shaped bird eye view (BEV) queries, generating more comprehensive and accurate BEV features. Additionally, we devise a cross-domain query adaptation as a heuristic to fuse 2D priors, implicitly encoding the candidate positions of targets. Furthermore, IFTR optimizes communication efficiency by sending instance-level features, achieving an optimal performance-bandwidth trade-off. We evaluate the proposed IFTR on a real dataset, DAIR-V2X, and two simulated datasets, OPV2V and V2XSet, achieving performance improvements of 57.96%, 9.23% and 12.99% in AP@70 metrics compared to the previous SOTAs, respectively. Extensive experiments demonstrate the superiority of IFTR and the effectiveness of its key components. The code is available at this https URL.
Title:
SeFi-CD: A Semantic First Change Detection Paradigm That Can Detect Any Change You Want
Authors: Ling Zhao, Zhenyang Huang, Dongsheng Kuang, Chengli Peng, Jun Gan, Haifeng Li
Abstract
The existing change detection(CD) methods can be summarized as the visual-first change detection (ViFi-CD) paradigm, which first extracts change features from visual differences and then assigns them specific semantic information. However, CD is essentially dependent on change regions of interest (CRoIs), meaning that the CD results are directly determined by the semantics changes of interest, making its primary image factor semantic of interest rather than visual. The ViFi-CD paradigm can only assign specific semantics of interest to specific change features extracted from visual differences, leading to the inevitable omission of potential CRoIs and the inability to adapt to different CRoI CD tasks. In other words, changes in other CRoIs cannot be detected by the ViFi-CD method without retraining the model or significantly modifying the method. This paper introduces a new CD paradigm, the semantic-first CD (SeFi-CD) paradigm. The core idea of SeFi-CD is to first perceive the dynamic semantics of interest and then visually search for change features related to the semantics. Based on the SeFi-CD paradigm, we designed Anything You Want Change Detection (AUWCD). Experiments on public datasets demonstrate that the AUWCD outperforms the current state-of-the-art CD methods, achieving an average F1 score 5.01\% higher than that of these advanced supervised baselines on the SECOND dataset, with a maximum increase of 13.17\%. The proposed SeFi-CD offers a novel CD perspective and approach.
Title:
Transferring Structure Knowledge: A New Task to Fake news Detection Towards Cold-Start Propagation
Authors: Lingwei Wei, Dou Hu, Wei Zhou, Songlin Hu
Subjects: Subjects:
Social and Information Networks (cs.SI); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
Abstract
Many fake news detection studies have achieved promising performance by extracting effective semantic and structure features from both content and propagation trees. However, it is challenging to apply them to practical situations, especially when using the trained propagation-based models to detect news with no propagation data. Towards this scenario, we study a new task named cold-start fake news detection, which aims to detect content-only samples with missing propagation. To achieve the task, we design a simple but effective Structure Adversarial Net (SAN) framework to learn transferable features from available propagation to boost the detection of content-only samples. SAN introduces a structure discriminator to estimate dissimilarities among learned features with and without propagation, and further learns structure-invariant features to enhance the generalization of existing propagation-based methods for content-only samples. We conduct qualitative and quantitative experiments on three datasets. Results show the challenge of the new task and the effectiveness of our SAN framework.
Title:
MutDet: Mutually Optimizing Pre-training for Remote Sensing Object Detection
Authors: Ziyue Huang, Yongchao Feng, Qingjie Liu, Yunhong Wang
Subjects: Subjects:
Computer Vision and Pattern Recognition (cs.CV)
Abstract
Detection pre-training methods for the DETR series detector have been extensively studied in natural scenes, e.g., DETReg. However, the detection pre-training remains unexplored in remote sensing scenes. In existing pre-training methods, alignment between object embeddings extracted from a pre-trained backbone and detector features is significant. However, due to differences in feature extraction methods, a pronounced feature discrepancy still exists and hinders the pre-training performance. The remote sensing images with complex environments and more densely distributed objects exacerbate the discrepancy. In this work, we propose a novel Mutually optimizing pre-training framework for remote sensing object Detection, dubbed as MutDet. In MutDet, we propose a systemic solution against this challenge. Firstly, we propose a mutual enhancement module, which fuses the object embeddings and detector features bidirectionally in the last encoder layer, enhancing their information interaction.Secondly, contrastive alignment loss is employed to guide this alignment process softly and simultaneously enhances detector features' discriminativity. Finally, we design an auxiliary siamese head to mitigate the task gap arising from the introduction of enhancement module. Comprehensive experiments on various settings show new state-of-the-art transfer performance. The improvement is particularly pronounced when data quantity is limited. When using 10% of the DIOR-R data, MutDet improves DetReg by 6.1% in AP50. Codes and models are available at: this https URL.
Title:
Optimizing ROI Benefits Vehicle ReID in ITS
Authors: Mei Qiu, Lauren Ann Christopher, Lingxi Li, Stanley Chien, Yaobin Chen
Subjects: Subjects:
Computer Vision and Pattern Recognition (cs.CV); Image and Video Processing (eess.IV)
Abstract
Vehicle re-identification (ReID) is a computer vision task that matches the same vehicle across different cameras or viewpoints in a surveillance system. This is crucial for Intelligent Transportation Systems (ITS), where the effectiveness is influenced by the regions from which vehicle images are cropped. This study explores whether optimal vehicle detection regions, guided by detection confidence scores, can enhance feature matching and ReID tasks. Using our framework with multiple Regions of Interest (ROIs) and lane-wise vehicle counts, we employed YOLOv8 for detection and DeepSORT for tracking across twelve Indiana Highway videos, including two pairs of videos from non-overlapping cameras. Tracked vehicle images were cropped from inside and outside the ROIs at five-frame intervals. Features were extracted using pre-trained models: ResNet50, ResNeXt50, Vision Transformer, and Swin-Transformer. Feature consistency was assessed through cosine similarity, information entropy, and clustering variance. Results showed that features from images cropped inside ROIs had higher mean cosine similarity values compared to those involving one image inside and one outside the ROIs. The most significant difference was observed during night conditions (0.7842 inside vs. 0.5 outside the ROI with Swin-Transformer) and in cross-camera scenarios (0.75 inside-inside vs. 0.52 inside-outside the ROI with Vision Transformer). Information entropy and clustering variance further supported that features in ROIs are more consistent. These findings suggest that strategically selected ROIs can enhance tracking performance and ReID accuracy in ITS.
Title:
Sim-to-Real Domain Adaptation for Deformation Classification
Authors: Joel Sol, Jamil Fayyad, Shadi Alijani, Homayoun Najjaran
Abstract
Deformation detection is vital for enabling accurate assessment and prediction of structural changes in materials, ensuring timely and effective interventions to maintain safety and integrity. Automating deformation detection through computer vision is crucial for efficient monitoring, but it faces significant challenges in creating a comprehensive dataset of both deformed and non-deformed objects, which can be difficult to obtain in many scenarios. In this paper, we introduce a novel framework for generating controlled synthetic data that simulates deformed objects. This approach allows for the realistic modeling of object deformations under various conditions. Our framework integrates an intelligent adapter network that facilitates sim-to-real domain adaptation, enhancing classification results without requiring real data from deformed objects. We conduct experiments on domain adaptation and classification tasks and demonstrate that our framework improves sim-to-real classification results compared to simulation baseline.
Title:
Reproducibility of Issues Reported in Stack Overflow Questions: Challenges, Impact & Estimation
Abstract
Software developers often submit questions to technical Q&A sites like Stack Overflow (SO) to resolve code-level problems. In practice, they include example code snippets with questions to explain the programming issues. Existing research suggests that users attempt to reproduce the reported issues using given code snippets when answering questions. Unfortunately, such code snippets could not always reproduce the issues due to several unmet challenges that prevent questions from receiving appropriate and prompt solutions. One previous study investigated reproducibility challenges and produced a catalog. However, how the practitioners perceive this challenge catalog is unknown. Practitioners' perspectives are inevitable in validating these challenges and estimating their severity. This study first surveyed 53 practitioners to understand their perspectives on reproducibility challenges. We attempt to (a) see whether they agree with these challenges, (b) determine the impact of each challenge on answering questions, and (c) identify the need for tools to promote reproducibility. Survey results show that - (a) about 90% of the participants agree with the challenges, (b) "missing an important part of code" most severely hurt reproducibility, and (c) participants strongly recommend introducing automated tool support to promote reproducibility. Second, we extract \emph{nine} code-based features (e.g., LOC, compilability) and build five Machine Learning (ML) models to predict issue reproducibility. Early detection might help users improve code snippets and their reproducibility. Our models achieve 84.5% precision, 83.0% recall, 82.8% F1-score, and 82.8% overall accuracy, which are highly promising. Third, we systematically interpret the ML model and explain how code snippets with reproducible issues differ from those with irreproducible issues.
Title:
Harnessing Feature Clustering For Enhanced Anomaly Detection With Variational Autoencoder And Dynamic Threshold
Authors: Tolulope Ale (1), Nicole-Jeanne Schlegel (2), Vandana P. Janeja (1) ((1) University of Maryland Baltimore County Baltimore MD USA, (2) National Oceanic and Atmospheric Administration Geophysical Fluid Dynamics Laboratory Princeton NJ USA)
Subjects: Subjects:
Machine Learning (cs.LG); Information Retrieval (cs.IR)
Abstract
We introduce an anomaly detection method for multivariate time series data with the aim of identifying critical periods and features influencing extreme climate events like snowmelt in the Arctic. This method leverages the Variational Autoencoder (VAE) integrated with dynamic thresholding and correlation-based feature clustering. This framework enhances the VAE's ability to identify localized dependencies and learn the temporal relationships in climate data, thereby improving the detection of anomalies as demonstrated by its higher F1-score on benchmark datasets. The study's main contributions include the development of a robust anomaly detection method, improving feature representation within VAEs through clustering, and creating a dynamic threshold algorithm for localized anomaly detection. This method offers explainability of climate anomalies across different regions.
Title:
Augmented Neural Fine-Tuning for Efficient Backdoor Purification
Authors: Nazmul Karim, Abdullah Al Arafat, Umar Khalid, Zhishan Guo, Nazanin Rahnavard
Subjects: Subjects:
Computer Vision and Pattern Recognition (cs.CV)
Abstract
Recent studies have revealed the vulnerability of deep neural networks (DNNs) to various backdoor attacks, where the behavior of DNNs can be compromised by utilizing certain types of triggers or poisoning mechanisms. State-of-the-art (SOTA) defenses employ too-sophisticated mechanisms that require either a computationally expensive adversarial search module for reverse-engineering the trigger distribution or an over-sensitive hyper-parameter selection module. Moreover, they offer sub-par performance in challenging scenarios, e.g., limited validation data and strong attacks. In this paper, we propose Neural mask Fine-Tuning (NFT) with an aim to optimally re-organize the neuron activities in a way that the effect of the backdoor is removed. Utilizing a simple data augmentation like MixUp, NFT relaxes the trigger synthesis process and eliminates the requirement of the adversarial search module. Our study further reveals that direct weight fine-tuning under limited validation data results in poor post-purification clean test accuracy, primarily due to overfitting issue. To overcome this, we propose to fine-tune neural masks instead of model weights. In addition, a mask regularizer has been devised to further mitigate the model drift during the purification process. The distinct characteristics of NFT render it highly efficient in both runtime and sample usage, as it can remove the backdoor even when a single sample is available from each class. We validate the effectiveness of NFT through extensive experiments covering the tasks of image classification, object detection, video action recognition, 3D point cloud, and natural language processing. We evaluate our method against 14 different attacks (LIRA, WaNet, etc.) on 11 benchmark data sets such as ImageNet, UCF101, Pascal VOC, ModelNet, OpenSubtitles2012, etc.
Title:
Plain-Det: A Plain Multi-Dataset Object Detector
Authors: Cheng Shi, Yuchen Zhu, Sibei Yang
Subjects: Subjects:
Computer Vision and Pattern Recognition (cs.CV)
Abstract
Recent advancements in large-scale foundational models have sparked widespread interest in training highly proficient large vision models. A common consensus revolves around the necessity of aggregating extensive, high-quality annotated data. However, given the inherent challenges in annotating dense tasks in computer vision, such as object detection and segmentation, a practical strategy is to combine and leverage all available data for training purposes. In this work, we propose Plain-Det, which offers flexibility to accommodate new datasets, robustness in performance across diverse datasets, training efficiency, and compatibility with various detection architectures. We utilize Def-DETR, with the assistance of Plain-Det, to achieve a mAP of 51.9 on COCO, matching the current state-of-the-art detectors. We conduct extensive experiments on 13 downstream datasets and Plain-Det demonstrates strong generalization capability. Code is release at this https URL
Title:
Constraints on Meso-Scale Structure in Complex Networks
Authors: Rudy Arthur
Subjects: Subjects:
Social and Information Networks (cs.SI); Physics and Society (physics.soc-ph)
Abstract
A key topic in network science is the detection of intermediate or meso-scale structures. Community, core-periphery, disassortative and other partitions allow us to understand the organisation and function of large networks. In this work we study under what conditions certain common meso-scale structures are detectable using the idea of block modularity. We find that the configuration model imposes strong restrictions on core-periphery and related structures in directed networks. We derive inequalities expressing when such structures can be detected under the configuration model. Nestedness is closely related to core-periphery and is similarly restricted to only be detectable under certain conditions. We show that these conditions are a generalisation of the resolution limit to structures other than assortative communities. We show how block modularity is related to the degree corrected Stochastic Block Model and that optimisation of one can be made equivalent to the other in general. Finally, we discuss these issues in inferential versus descriptive approaches to meso-scale structure detection.
Title:
When Pedestrian Detection Meets Multi-Modal Learning: Generalist Model and Benchmark Dataset
Authors: Yi Zhang, Wang Zeng, Sheng Jin, Chen Qian, Ping Luo, Wentao Liu
Subjects: Subjects:
Computer Vision and Pattern Recognition (cs.CV)
Abstract
Recent years have witnessed increasing research attention towards pedestrian detection by taking the advantages of different sensor modalities (e.g. RGB, IR, Depth, LiDAR and Event). However, designing a unified generalist model that can effectively process diverse sensor modalities remains a challenge. This paper introduces MMPedestron, a novel generalist model for multimodal perception. Unlike previous specialist models that only process one or a pair of specific modality inputs, MMPedestron is able to process multiple modal inputs and their dynamic combinations. The proposed approach comprises a unified encoder for modal representation and fusion and a general head for pedestrian detection. We introduce two extra learnable tokens, i.e. MAA and MAF, for adaptive multi-modal feature fusion. In addition, we construct the MMPD dataset, the first large-scale benchmark for multi-modal pedestrian detection. This benchmark incorporates existing public datasets and a newly collected dataset called EventPed, covering a wide range of sensor modalities including RGB, IR, Depth, LiDAR, and Event data. With multi-modal joint training, our model achieves state-of-the-art performance on a wide range of pedestrian detection benchmarks, surpassing leading models tailored for specific sensor modality. For example, it achieves 71.1 AP on COCO-Persons and 72.6 AP on LLVIP. Notably, our model achieves comparable performance to the InternImage-H model on CrowdHuman with 30x smaller parameters. Codes and data are available at this https URL.
Title:
FSD-BEV: Foreground Self-Distillation for Multi-view 3D Object Detection
Abstract
Although multi-view 3D object detection based on the Bird's-Eye-View (BEV) paradigm has garnered widespread attention as an economical and deployment-friendly perception solution for autonomous driving, there is still a performance gap compared to LiDAR-based methods. In recent years, several cross-modal distillation methods have been proposed to transfer beneficial information from teacher models to student models, with the aim of enhancing performance. However, these methods face challenges due to discrepancies in feature distribution originating from different data modalities and network structures, making knowledge transfer exceptionally challenging. In this paper, we propose a Foreground Self-Distillation (FSD) scheme that effectively avoids the issue of distribution discrepancies, maintaining remarkable distillation effects without the need for pre-trained teacher models or cumbersome distillation strategies. Additionally, we design two Point Cloud Intensification (PCI) strategies to compensate for the sparsity of point clouds by frame combination and pseudo point assignment. Finally, we develop a Multi-Scale Foreground Enhancement (MSFE) module to extract and fuse multi-scale foreground features by predicted elliptical Gaussian heatmap, further improving the model's performance. We integrate all the above innovations into a unified framework named FSD-BEV. Extensive experiments on the nuScenes dataset exhibit that FSD-BEV achieves state-of-the-art performance, highlighting its effectiveness. The code and models are available at: this https URL.
Title:
Lost and Found: Overcoming Detector Failures in Online Multi-Object Tracking
Authors: Lorenzo Vaquero, Yihong Xu, Xavier Alameda-Pineda, Victor M. Brea, Manuel Mucientes
Subjects: Subjects:
Computer Vision and Pattern Recognition (cs.CV)
Abstract
Multi-object tracking (MOT) endeavors to precisely estimate the positions and identities of multiple objects over time. The prevailing approach, tracking-by-detection (TbD), first detects objects and then links detections, resulting in a simple yet effective method. However, contemporary detectors may occasionally miss some objects in certain frames, causing trackers to cease tracking prematurely. To tackle this issue, we propose BUSCA, meaning `to search', a versatile framework compatible with any online TbD system, enhancing its ability to persistently track those objects missed by the detector, primarily due to occlusions. Remarkably, this is accomplished without modifying past tracking results or accessing future frames, i.e., in a fully online manner. BUSCA generates proposals based on neighboring tracks, motion, and learned tokens. Utilizing a decision Transformer that integrates multimodal visual and spatiotemporal information, it addresses the object-proposal association as a multi-choice question-answering task. BUSCA is trained independently of the underlying tracker, solely on synthetic data, without requiring fine-tuning. Through BUSCA, we showcase consistent performance enhancements across five different trackers and establish a new state-of-the-art baseline across three different benchmarks. Code available at: this https URL.
Title:
LabelDistill: Label-guided Cross-modal Knowledge Distillation for Camera-based 3D Object Detection
Authors: Sanmin Kim, Youngseok Kim, Sihwan Hwang, Hyeonjun Jeong, Dongsuk Kum
Subjects: Subjects:
Computer Vision and Pattern Recognition (cs.CV)
Abstract
Recent advancements in camera-based 3D object detection have introduced cross-modal knowledge distillation to bridge the performance gap with LiDAR 3D detectors, leveraging the precise geometric information in LiDAR point clouds. However, existing cross-modal knowledge distillation methods tend to overlook the inherent imperfections of LiDAR, such as the ambiguity of measurements on distant or occluded objects, which should not be transferred to the image detector. To mitigate these imperfections in LiDAR teacher, we propose a novel method that leverages aleatoric uncertainty-free features from ground truth labels. In contrast to conventional label guidance approaches, we approximate the inverse function of the teacher's head to effectively embed label inputs into feature space. This approach provides additional accurate guidance alongside LiDAR teacher, thereby boosting the performance of the image detector. Additionally, we introduce feature partitioning, which effectively transfers knowledge from the teacher modality while preserving the distinctive features of the student, thereby maximizing the potential of both modalities. Experimental results demonstrate that our approach improves mAP and NDS by 5.1 points and 4.9 points compared to the baseline model, proving the effectiveness of our approach. The code is available at this https URL
Title:
Few-Shot Bioacoustic Event Detection with Frame-Level Embedding Learning System
Authors: PengYuan Zhao, ChengWei Lu, Liang Zou
Subjects: Subjects:
Sound (cs.SD); Audio and Speech Processing (eess.AS)
Abstract
This technical report presents our frame-level embedding learning system for the DCASE2024 challenge for few-shot bioacoustic event detection (Task 5).In this work, we used log-mel and PCEN for feature extraction of the input audio, Netmamba Encoder as the information interaction network, and adopted data augmentation strategies to improve the generalizability of the trained model as well as multiple post-processing methods. Our final system achieved an F-measure score of 56.4%, securing the 2nd rank in the few-shot bioacoustic event detection category of the Detection and Classification of Acoustic Scenes and Events Challenge 2024.
Title:
V2I-Calib: A Novel Calibration Approach for Collaborative Vehicle and Infrastructure LiDAR Systems
Abstract
Cooperative vehicle and infrastructure LiDAR systems hold great potential, yet their implementation faces numerous challenges. Calibration of LiDAR systems across heterogeneous vehicle and infrastructure endpoints is a critical step to ensure the accuracy and consistency of perception system data, necessitating calibration methods that are real-time and stable. To this end, this paper introduces a novel calibration method for cooperative vehicle and road infrastructure LiDAR systems, which exploits spatial association information between detection boxes. The method centers around a novel Overall IoU metric that reflects the correlation of targets between vehicle and infrastructure, enabling real-time monitoring of calibration results. We search for common matching boxes between vehicle and infrastructure nodes by constructing an affinity matrix. Subsequently, these matching boxes undergo extrinsic parameter computation and optimization. Comparative and ablation experiments on the DAIR-V2X dataset confirm the superiority of our method. To better reflect the differences in calibration results, we have categorized the calibration tasks on the DAIR-V2X dataset based on their level of difficulty, enriching the dataset's utility for future research. Our project is available at this https URL .
Title:
Shape2Scene: 3D Scene Representation Learning Through Pre-training on Shape Data
Authors: Tuo Feng, Wenguan Wang, Ruijie Quan, Yi Yang
Abstract
Current 3D self-supervised learning methods of 3D scenes face a data desert issue, resulting from the time-consuming and expensive collecting process of 3D scene data. Conversely, 3D shape datasets are easier to collect. Despite this, existing pre-training strategies on shape data offer limited potential for 3D scene understanding due to significant disparities in point quantities. To tackle these challenges, we propose Shape2Scene (S2S), a novel method that learns representations of large-scale 3D scenes from 3D shape data. We first design multiscale and high-resolution backbones for shape and scene level 3D tasks, i.e., MH-P (point-based) and MH-V (voxel-based). MH-P/V establishes direct paths to highresolution features that capture deep semantic information across multiple scales. This pivotal nature makes them suitable for a wide range of 3D downstream tasks that tightly rely on high-resolution features. We then employ a Shape-to-Scene strategy (S2SS) to amalgamate points from various shapes, creating a random pseudo scene (comprising multiple objects) for training data, mitigating disparities between shapes and scenes. Finally, a point-point contrastive loss (PPC) is applied for the pre-training of MH-P/V. In PPC, the inherent correspondence (i.e., point pairs) is naturally obtained in S2SS. Extensive experiments have demonstrated the transferability of 3D representations learned by MH-P/V across shape-level and scene-level 3D tasks. MH-P achieves notable performance on well-known point cloud datasets (93.8% OA on ScanObjectNN and 87.6% instance mIoU on ShapeNetPart). MH-V also achieves promising performance in 3D semantic segmentation and 3D object detection.
Title:
Efficient Facial Landmark Detection for Embedded Systems
Authors: Ji-Jia Wu
Subjects: Subjects:
Computer Vision and Pattern Recognition (cs.CV)
Abstract
This paper introduces the Efficient Facial Landmark Detection (EFLD) model, specifically designed for edge devices confronted with the challenges related to power consumption and time latency. EFLD features a lightweight backbone and a flexible detection head, each significantly enhancing operational efficiency on resource-constrained devices. To improve the model's robustness, we propose a cross-format training strategy. This strategy leverages a wide variety of publicly accessible datasets to enhance the model's generalizability and robustness, without increasing inference costs. Our ablation study highlights the significant impact of each component on reducing computational demands, model size, and improving accuracy. EFLD demonstrates superior performance compared to competitors in the IEEE ICME 2024 Grand Challenges PAIR Competition, a contest focused on low-power, efficient, and accurate facial-landmark detection for embedded systems, showcasing its effectiveness in real-world facial landmark detection tasks.
Title:
Follow the Rules: Reasoning for Video Anomaly Detection with Large Language Models
Abstract
Video Anomaly Detection (VAD) is crucial for applications such as security surveillance and autonomous driving. However, existing VAD methods provide little rationale behind detection, hindering public trust in real-world deployments. In this paper, we approach VAD with a reasoning framework. Although Large Language Models (LLMs) have shown revolutionary reasoning ability, we find that their direct use falls short of VAD. Specifically, the implicit knowledge pre-trained in LLMs focuses on general context and thus may not apply to every specific real-world VAD scenario, leading to inflexibility and inaccuracy. To address this, we propose AnomalyRuler, a novel rule-based reasoning framework for VAD with LLMs. AnomalyRuler comprises two main stages: induction and deduction. In the induction stage, the LLM is fed with few-shot normal reference samples and then summarizes these normal patterns to induce a set of rules for detecting anomalies. The deduction stage follows the induced rules to spot anomalous frames in test videos. Additionally, we design rule aggregation, perception smoothing, and robust reasoning strategies to further enhance AnomalyRuler's robustness. AnomalyRuler is the first reasoning approach for the one-class VAD task, which requires only few-normal-shot prompting without the need for full-shot training, thereby enabling fast adaption to various VAD scenarios. Comprehensive experiments across four VAD benchmarks demonstrate AnomalyRuler's state-of-the-art detection performance and reasoning ability.
Title:
Exploring the Impact of Moire Pattern on Deepfake Detectors
Authors: Razaib Tariq, Shahroz Tariq, Simon S. Woo
Subjects: Subjects:
Computer Vision and Pattern Recognition (cs.CV)
Abstract
Deepfake detection is critical in mitigating the societal threats posed by manipulated videos. While various algorithms have been developed for this purpose, challenges arise when detectors operate externally, such as on smartphones, when users take a photo of deepfake images and upload on the Internet. One significant challenge in such scenarios is the presence of Moiré patterns, which degrade image quality and confound conventional classification algorithms, including deep neural networks (DNNs). The impact of Moiré patterns remains largely unexplored for deepfake detectors. In this study, we investigate how camera-captured deepfake videos from digital screens affect detector performance. We conducted experiments using two prominent datasets, CelebDF and FF++, comparing the performance of four state-of-the-art detectors on camera-captured deepfake videos with introduced Moiré patterns. Our findings reveal a significant decline in detector accuracy, with none achieving above 68% on average. This underscores the critical need to address Moiré pattern challenges in real-world deepfake detection scenarios.
Title:
Melon Fruit Detection and Quality Assessment Using Generative AI-Based Image Data Augmentation
Abstract
Monitoring and managing the growth and quality of fruits are very important tasks. To effectively train deep learning models like YOLO for real-time fruit detection, high-quality image datasets are essential. However, such datasets are often lacking in agriculture. Generative AI models can help create high-quality images. In this study, we used MidJourney and Firefly tools to generate images of melon greenhouses and post-harvest fruits through text-to-image, pre-harvest image-to-image, and post-harvest image-to-image methods. We evaluated these AIgenerated images using PSNR and SSIM metrics and tested the detection performance of the YOLOv9 model. We also assessed the net quality of real and generated fruits. Our results showed that generative AI could produce images very similar to real ones, especially for post-harvest fruits. The YOLOv9 model detected the generated images well, and the net quality was also measurable. This shows that generative AI can create realistic images useful for fruit detection and quality assessment, indicating its great potential in agriculture. This study highlights the potential of AI-generated images for data augmentation in melon fruit detection and quality assessment and envisions a positive future for generative AI applications in agriculture.
Title:
Omni-Dimensional Frequency Learner for General Time Series Analysis
Abstract
Frequency domain representation of time series feature offers a concise representation for handling real-world time series data with inherent complexity and dynamic nature. However, current frequency-based methods with complex operations still fall short of state-of-the-art time domain methods for general time series analysis. In this work, we present Omni-Dimensional Frequency Learner (ODFL) model based on a in depth analysis among all the three aspects of the spectrum feature: channel redundancy property among the frequency dimension, the sparse and un-salient frequency energy distribution among the frequency dimension, and the semantic diversity among the variable dimension. Technically, our method is composed of a semantic-adaptive global filter with attention to the un-salient frequency bands and partial operation among the channel dimension. Empirical results show that ODFL achieves consistent state-of-the-art in five mainstream time series analysis tasks, including short- and long-term forecasting, imputation, classification, and anomaly detection, offering a promising foundation for time series analysis.
Abstract
Recently, deep learning-based Image-to-Image (I2I) networks have become the predominant choice for I2I tasks such as image super-resolution and denoising. Despite their remarkable performance, the backdoor vulnerability of I2I networks has not been explored. To fill this research gap, we conduct a comprehensive investigation on the susceptibility of I2I networks to backdoor attacks. Specifically, we propose a novel backdoor attack technique, where the compromised I2I network behaves normally on clean input images, yet outputs a predefined image of the adversary for malicious input images containing the trigger. To achieve this I2I backdoor attack, we propose a targeted universal adversarial perturbation (UAP) generation algorithm for I2I networks, where the generated UAP is used as the backdoor trigger. Additionally, in the backdoor training process that contains the main task and the backdoor task, multi-task learning (MTL) with dynamic weighting methods is employed to accelerate convergence rates. In addition to attacking I2I tasks, we extend our I2I backdoor to attack downstream tasks, including image classification and object detection. Extensive experiments demonstrate the effectiveness of the I2I backdoor on state-of-the-art I2I network architectures, as well as the robustness against different mainstream backdoor defenses.
Title:
GROOT: Generating Robust Watermark for Diffusion-Model-Based Audio Synthesis
Abstract
Amid the burgeoning development of generative models like diffusion models, the task of differentiating synthesized audio from its natural counterpart grows more daunting. Deepfake detection offers a viable solution to combat this challenge. Yet, this defensive measure unintentionally fuels the continued refinement of generative models. Watermarking emerges as a proactive and sustainable tactic, preemptively regulating the creation and dissemination of synthesized content. Thus, this paper, as a pioneer, proposes the generative robust audio watermarking method (Groot), presenting a paradigm for proactively supervising the synthesized audio and its source diffusion models. In this paradigm, the processes of watermark generation and audio synthesis occur simultaneously, facilitated by parameter-fixed diffusion models equipped with a dedicated encoder. The watermark embedded within the audio can subsequently be retrieved by a lightweight decoder. The experimental results highlight Groot's outstanding performance, particularly in terms of robustness, surpassing that of the leading state-of-the-art methods. Beyond its impressive resilience against individual post-processing attacks, Groot exhibits exceptional robustness when facing compound attacks, maintaining an average watermark extraction accuracy of around 95%.
Title:
Effective Motion Modeling for UAV-platform Multiple Object Tracking with Re-Margin Loss
Authors: Mufeng Yao, Jinlong Peng, Qingdong He, Bo Peng, Hao Chen, Mingmin Chi, Chao Liu, Jon Atli Benediktsson
Subjects: Subjects:
Computer Vision and Pattern Recognition (cs.CV)
Abstract
Multiple object tracking (MOT) from unmanned aerial vehicle (UAV) platforms requires efficient motion modeling. This is because UAV-MOT faces tracking difficulties caused by large and irregular motion, and insufficient training due to the motion long-tailed distribution of current UAV-MOT datasets. Previous UAV-MOT methods either extract motion and detection features redundantly or supervise motion model in a sparse scheme, which limited their tracking performance and speed. To this end, we propose a flowing-by-detection module to realize accurate motion modeling with a minimum cost. Focusing on the motion long-tailed problem that were ignored by previous works, the flow-guided margin loss is designed to enable more complete training of large moving objects. Experiments on two widely open-source datasets show that our proposed model can successfully track objects with large and irregular motion and outperform existing state-of-the-art methods in UAV-MOT tasks.
Title:
A Survey of Defenses against AI-generated Visual Media: Detection, Disruption, and Authentication
Abstract
Deep generative models have demonstrated impressive performance in various computer vision applications, including image synthesis, video generation, and medical analysis. Despite their significant advancements, these models may be used for malicious purposes, such as misinformation, deception, and copyright violation. In this paper, we provide a systematic and timely review of research efforts on defenses against AI-generated visual media, covering detection, disruption, and authentication. We review existing methods and summarize the mainstream defense-related tasks within a unified passive and proactive framework. Moreover, we survey the derivative tasks concerning the trustworthiness of defenses, such as their robustness and fairness. For each task, we formulate its general pipeline and propose a taxonomy based on methodological strategies that are uniformly applicable to the primary subtasks. Additionally, we summarize the commonly used evaluation datasets, criteria, and metrics. Finally, by analyzing the reviewed studies, we provide insights into current research challenges and suggest possible directions for future research.
Abstract
Open-vocabulary object detection focusing on detecting novel categories guided by natural language. In this report, we propose Open-Vocabulary Light-Weighted Detection Transformer (OVLW-DETR), a deployment friendly open-vocabulary detector with strong performance and low latency. Building upon OVLW-DETR, we provide an end-to-end training recipe that transferring knowledge from vision-language model (VLM) to object detector with simple alignment. We align detector with the text encoder from VLM by replacing the fixed classification layer weights in detector with the class-name embeddings extracted from the text encoder. Without additional fusing module, OVLW-DETR is flexible and deployment friendly, making it easier to implement and modulate. improving the efficiency of interleaved attention computation. Experimental results demonstrate that the proposed approach is superior over existing real-time open-vocabulary detectors on standard Zero-Shot LVIS benchmark. Source code and pre-trained models are available at [this https URL].
Title:
Anticipating Future Object Compositions without Forgetting
Abstract
Despite the significant advancements in computer vision models, their ability to generalize to novel object-attribute compositions remains limited. Existing methods for Compositional Zero-Shot Learning (CZSL) mainly focus on image classification. This paper aims to enhance CZSL in object detection without forgetting prior learned knowledge. We use Grounding DINO and incorporate Compositional Soft Prompting (CSP) into it and extend it with Compositional Anticipation. We achieve a 70.5% improvement over CSP on the harmonic mean (HM) between seen and unseen compositions on the CLEVR dataset. Furthermore, we introduce Contrastive Prompt Tuning to incrementally address model confusion between similar compositions. We demonstrate the effectiveness of this method and achieve an increase of 14.5% in HM across the pretrain, increment, and unseen sets. Collectively, these methods provide a framework for learning various compositions with limited data, as well as improving the performance of underperforming compositions when additional data becomes available.
Title:
When Synthetic Traces Hide Real Content: Analysis of Stable Diffusion Image Laundering
Authors: Sara Mandelli, Paolo Bestagini, Stefano Tubaro
Abstract
In recent years, methods for producing highly realistic synthetic images have significantly advanced, allowing the creation of high-quality images from text prompts that describe the desired content. Even more impressively, Stable Diffusion (SD) models now provide users with the option of creating synthetic images in an image-to-image translation fashion, modifying images in the latent space of advanced autoencoders. This striking evolution, however, brings an alarming consequence: it is possible to pass an image through SD autoencoders to reproduce a synthetic copy of the image with high realism and almost no visual artifacts. This process, known as SD image laundering, can transform real images into lookalike synthetic ones and risks complicating forensic analysis for content authenticity verification. Our paper investigates the forensic implications of image laundering, revealing a serious potential to obscure traces of real content, including sensitive and harmful materials that could be mistakenly classified as synthetic, thereby undermining the protection of individuals depicted. To address this issue, we propose a two-stage detection pipeline that effectively differentiates between pristine, laundered, and fully synthetic images (those generated from text prompts), showing robustness across various conditions. Finally, we highlight another alarming property of image laundering, which appears to mask the unique artifacts exploited by forensic detectors to solve the camera model identification task, strongly undermining their performance. Our experimental code is available at this https URL.
Title:
SEED: A Simple and Effective 3D DETR in Point Clouds
Authors: Zhe Liu, Jinghua Hou, Xiaoqing Ye, Tong Wang, Jingdong Wang, Xiang Bai
Subjects: Subjects:
Computer Vision and Pattern Recognition (cs.CV)
Abstract
Recently, detection transformers (DETRs) have gradually taken a dominant position in 2D detection thanks to their elegant framework. However, DETR-based detectors for 3D point clouds are still difficult to achieve satisfactory performance. We argue that the main challenges are twofold: 1) How to obtain the appropriate object queries is challenging due to the high sparsity and uneven distribution of point clouds; 2) How to implement an effective query interaction by exploiting the rich geometric structure of point clouds is not fully explored. To this end, we propose a simple and effective 3D DETR method (SEED) for detecting 3D objects from point clouds, which involves a dual query selection (DQS) module and a deformable grid attention (DGA) module. More concretely, to obtain appropriate queries, DQS first ensures a high recall to retain a large number of queries by the predicted confidence scores and then further picks out high-quality queries according to the estimated quality scores. DGA uniformly divides each reference box into grids as the reference points and then utilizes the predicted offsets to achieve a flexible receptive field, allowing the network to focus on relevant regions and capture more informative features. Extensive ablation studies on DQS and DGA demonstrate its effectiveness. Furthermore, our SEED achieves state-of-the-art detection performance on both the large-scale Waymo and nuScenes datasets, illustrating the superiority of our proposed method. The code is available at this https URL
Title:
OPEN: Object-wise Position Embedding for Multi-view 3D Object Detection
Authors: Jinghua Hou, Tong Wang, Xiaoqing Ye, Zhe Liu, Shi Gong, Xiao Tan, Errui Ding, Jingdong Wang, Xiang Bai
Subjects: Subjects:
Computer Vision and Pattern Recognition (cs.CV)
Abstract
Accurate depth information is crucial for enhancing the performance of multi-view 3D object detection. Despite the success of some existing multi-view 3D detectors utilizing pixel-wise depth supervision, they overlook two significant phenomena: 1) the depth supervision obtained from LiDAR points is usually distributed on the surface of the object, which is not so friendly to existing DETR-based 3D detectors due to the lack of the depth of 3D object center; 2) for distant objects, fine-grained depth estimation of the whole object is more challenging. Therefore, we argue that the object-wise depth (or 3D center of the object) is essential for accurate detection. In this paper, we propose a new multi-view 3D object detector named OPEN, whose main idea is to effectively inject object-wise depth information into the network through our proposed object-wise position embedding. Specifically, we first employ an object-wise depth encoder, which takes the pixel-wise depth map as a prior, to accurately estimate the object-wise depth. Then, we utilize the proposed object-wise position embedding to encode the object-wise depth information into the transformer decoder, thereby producing 3D object-aware features for final detection. Extensive experiments verify the effectiveness of our proposed method. Furthermore, OPEN achieves a new state-of-the-art performance with 64.4% NDS and 56.7% mAP on the nuScenes test benchmark.
Title:
An Autonomous Drone Swarm for Detecting and Tracking Anomalies among Dense Vegetation
Authors: Rakesh John Amala Arokia Nathan, Sigrid Strand, Daniel Mehrwald, Dmitriy Shutin, Oliver Bimber
Subjects: Subjects:
Computer Vision and Pattern Recognition (cs.CV)
Abstract
Swarms of drones offer an increased sensing aperture, and having them mimic behaviors of natural swarms enhances sampling by adapting the aperture to local conditions. We demonstrate that such an approach makes detecting and tracking heavily occluded targets practically feasible. While object classification applied to conventional aerial images generalizes poorly the randomness of occlusion and is therefore inefficient even under lightly occluded conditions, anomaly detection applied to synthetic aperture integral images is robust for dense vegetation, such as forests, and is independent of pre-trained classes. Our autonomous swarm searches the environment for occurrences of the unknown or unexpected, tracking them while continuously adapting its sampling pattern to optimize for local viewing conditions. In our real-life field experiments with a swarm of six drones, we achieved an average positional accuracy of 0.39 m with an average precision of 93.2% and an average recall of 95.9%. Here, adapted particle swarm optimization considers detection confidences and predicted target appearance. We show that sensor noise can effectively be included in the synthetic aperture image integration process, removing the need for a computationally costly optimization of high-dimensional parameter spaces. Finally, we present a complete hard- and software framework that supports low-latency transmission (approx. 80 ms round-trip time) and fast processing (approx. 600 ms per formation step) of extensive (70-120 Mbit/s) video and telemetry data, and swarm control for swarms of up to ten drones.
Title:
FabGPT: An Efficient Large Multimodal Model for Complex Wafer Defect Knowledge Queries
Abstract
Intelligence is key to advancing integrated circuit (IC) fabrication. Recent breakthroughs in Large Multimodal Models (LMMs) have unlocked unparalleled abilities in understanding images and text, fostering intelligent fabrication. Leveraging the power of LMMs, we introduce FabGPT, a customized IC fabrication large multimodal model for wafer defect knowledge query. FabGPT manifests expertise in conducting defect detection in Scanning Electron Microscope (SEM) images, performing root cause analysis, and providing expert question-answering (Q&A) on fabrication processes. FabGPT matches enhanced multimodal features to automatically detect minute defects under complex wafer backgrounds and reduce the subjectivity of manual threshold settings. Besides, the proposed modulation module and interactive corpus training strategy embed wafer defect knowledge into the pre-trained model, effectively balancing Q&A queries related to defect knowledge and original knowledge and mitigating the modality bias issues. Experiments on in-house fab data (SEM-WaD) show that our FabGPT achieves significant performance improvement in wafer defect detection and knowledge querying.
Title:
BiasScanner: Automatic Detection and Classification of News Bias to Strengthen Democracy
Authors: Tim Menzner, Jochen L. Leidner
Subjects: Subjects:
Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Information Retrieval (cs.IR)
Abstract
The increasing consumption of news online in the 21st century coincided with increased publication of disinformation, biased reporting, hate speech and other unwanted Web content. We describe BiasScanner, an application that aims to strengthen democracy by supporting news consumers with scrutinizing news articles they are reading online. BiasScanner contains a server-side pre-trained large language model to identify biased sentences of news articles and a front-end Web browser plug-in. At the time of writing, BiasScanner can identify and classify more than two dozen types of media bias at the sentence level, making it the most fine-grained model and only deployed application (automatic system in use) of its kind. It was implemented in a light-weight and privacy-respecting manner, and in addition to highlighting likely biased sentence it also provides explanations for each classification decision as well as a summary analysis for each news article. While prior research has addressed news bias detection, we are not aware of any work that resulted in a deployed browser plug-in (c.f. also this http URL for a Web demo).
Title:
R3D-AD: Reconstruction via Diffusion for 3D Anomaly Detection
Abstract
3D anomaly detection plays a crucial role in monitoring parts for localized inherent defects in precision manufacturing. Embedding-based and reconstruction-based approaches are among the most popular and successful methods. However, there are two major challenges to the practical application of the current approaches: 1) the embedded models suffer the prohibitive computational and storage due to the memory bank structure; 2) the reconstructive models based on the MAE mechanism fail to detect anomalies in the unmasked regions. In this paper, we propose R3D-AD, reconstructing anomalous point clouds by diffusion model for precise 3D anomaly detection. Our approach capitalizes on the data distribution conversion of the diffusion process to entirely obscure the input's anomalous geometry. It step-wisely learns a strict point-level displacement behavior, which methodically corrects the aberrant points. To increase the generalization of the model, we further present a novel 3D anomaly simulation strategy named Patch-Gen to generate realistic and diverse defect shapes, which narrows the domain gap between training and testing. Our R3D-AD ensures a uniform spatial transformation, which allows straightforwardly generating anomaly results by distance comparison. Extensive experiments show that our R3D-AD outperforms previous state-of-the-art methods, achieving 73.4% Image-level AUROC on the Real3D-AD dataset and 74.9% Image-level AUROC on the Anomaly-ShapeNet dataset with an exceptional efficiency.
Title:
RepVF: A Unified Vector Fields Representation for Multi-task 3D Perception
Abstract
Concurrent processing of multiple autonomous driving 3D perception tasks within the same spatiotemporal scene poses a significant challenge, in particular due to the computational inefficiencies and feature competition between tasks when using traditional multi-task learning approaches. This paper addresses these issues by proposing a novel unified representation, RepVF, which harmonizes the representation of various perception tasks such as 3D object detection and 3D lane detection within a single framework. RepVF characterizes the structure of different targets in the scene through a vector field, enabling a single-head, multi-task learning model that significantly reduces computational redundancy and feature competition. Building upon RepVF, we introduce RFTR, a network designed to exploit the inherent connections between different tasks by utilizing a hierarchical structure of queries that implicitly model the relationships both between and within tasks. This approach eliminates the need for task-specific heads and parameters, fundamentally reducing the conflicts inherent in traditional multi-task learning paradigms. We validate our approach by combining labels from the OpenLane dataset with the Waymo Open dataset. Our work presents a significant advancement in the efficiency and effectiveness of multi-task perception in autonomous driving, offering a new perspective on handling multiple 3D perception tasks synchronously and in parallel. The code will be available at: this https URL
Title:
A BlueROV2-based platform for underwater mapping experiments
Abstract
We propose a low-cost laboratory platform for development and validation of underwater mapping techniques, using the BlueROV2 Remotely Operated Vehicle (ROV). Both the ROV and the objects to be mapped are placed in a pool that is imaged via an overhead camera. In our prototype mapping application, the ROV's pose is found using extended Kalman filtering on measurements from the overhead camera, inertial, and pressure sensors; while objects are detected with a deep neural network in the ROV camera stream. Validation experiments are performed for pose estimation, detection, and mapping. The litter detection dataset and code are made publicly available.
Title:
Interpreting Hand gestures using Object Detection and Digits Classification
Abstract
Hand gestures have evolved into a natural and intuitive means of engaging with technology. The objective of this research is to develop a robust system that can accurately recognize and classify hand gestures representing numbers. The proposed approach involves collecting a dataset of hand gesture images, preprocessing and enhancing the images, extracting relevant features, and training a machine learning model. The advancement of computer vision technology and object detection techniques, in conjunction with OpenCV's capability to analyze and comprehend hand gestures, presents a chance to transform the identification of numerical digits and its potential applications. The advancement of computer vision technology and object identification technologies, along with OpenCV's capacity to analyze and interpret hand gestures, has the potential to revolutionize human interaction, boosting people's access to information, education, and employment opportunities. Keywords: Computer Vision, Machine learning, Deep Learning, Neural Networks
Title:
A unified theory and statistical learning approach for traffic conflict detection
Authors: Yiru Jiao, Simeon C. Calvert, Sander van Cranenburgh, Hans van Lint
Abstract
This study proposes a unified theory and statistical learning approach for traffic conflict detection, addressing the long-existing call for a consistent and comprehensive methodology to evaluate the collision risk emerged in road user interactions. The proposed theory assumes a context-dependent probabilistic collision risk and frames conflict detection as estimating the risk by statistical learning from observed proximities and contextual variables. Three primary tasks are integrated: representing interaction context from selected observables, inferring proximity distributions in different contexts, and applying extreme value theory to relate conflict intensity with conflict probability. As a result, this methodology is adaptable to various road users and interaction scenarios, enhancing its applicability without the need for pre-labelled conflict data. Demonstration experiments are executed using real-world trajectory data, with the unified metric trained on lane-changing interactions on German highways and applied to near-crash events from the 100-Car Naturalistic Driving Study in the U.S. The experiments demonstrate the methodology's ability to provide effective collision warnings, generalise across different datasets and traffic environments, cover a broad range of conflicts, and deliver a long-tailed distribution of conflict intensity. This study contributes to traffic safety by offering a consistent and explainable methodology for conflict detection applicable across various scenarios. Its societal implications include enhanced safety evaluations of traffic infrastructures, more effective collision warning systems for autonomous and driving assistance systems, and a deeper understanding of road user behaviour in different traffic conditions, contributing to a potential reduction in accident rates and improving overall traffic safety.
Keyword: face recognition
Title:
Exploring Thermography Technology: A Comprehensive Facial Dataset for Face Detection, Recognition, and Emotion
Authors: Mohamed Fawzi Abdelshafie Abuhussein, Ashraf Darwish, Aboul Ella Hassanien
Abstract
This dataset includes 6823 thermal images captured using a UNI-T UTi165A camera for face detection, recognition, and emotion analysis. It consists of 2485 facial recognition images depicting emotions (happy, sad, angry, natural, surprised), 2054 images for face recognition, and 2284 images for face detection. The dataset covers various conditions, color palettes, shooting angles, and zoom levels, with a temperature range of -10°C to 400°C and a resolution of 19,200 pixels. It serves as a valuable resource for advancing thermal imaging technology, aiding in algorithm development, and benchmarking for facial recognition across different palettes. Additionally, it contributes to facial motion recognition, fostering interdisciplinary collaboration in computer vision, psychology, and neuroscience. The dataset promotes transparency in thermal face detection and recognition research, with applications in security, healthcare, and human-computer interaction.
Title:
Wicked Oddities: Selectively Poisoning for Effective Clean-Label Backdoor Attacks
Authors: Quang H. Nguyen, Nguyen Ngoc-Hieu, The-Anh Ta, Thanh Nguyen-Tang, Hoang Thanh-Tung, Khoa D. Doan
Subjects: Subjects:
Machine Learning (cs.LG); Cryptography and Security (cs.CR); Computer Vision and Pattern Recognition (cs.CV)
Abstract
Deep neural networks are vulnerable to backdoor attacks, a type of adversarial attack that poisons the training data to manipulate the behavior of models trained on such data. Clean-label attacks are a more stealthy form of backdoor attacks that can perform the attack without changing the labels of poisoned data. Early works on clean-label attacks added triggers to a random subset of the training set, ignoring the fact that samples contribute unequally to the attack's success. This results in high poisoning rates and low attack success rates. To alleviate the problem, several supervised learning-based sample selection strategies have been proposed. However, these methods assume access to the entire labeled training set and require training, which is expensive and may not always be practical. This work studies a new and more practical (but also more challenging) threat model where the attacker only provides data for the target class (e.g., in face recognition systems) and has no knowledge of the victim model or any other classes in the training set. We study different strategies for selectively poisoning a small set of training samples in the target class to boost the attack success rate in this setting. Our threat model poses a serious threat in training machine learning models with third-party datasets, since the attack can be performed effectively with limited information. Experiments on benchmark datasets illustrate the effectiveness of our strategies in improving clean-label backdoor attacks.
Keyword: augmentation
Title:
Representation Debiasing of Generated Data Involving Domain Experts
Abstract
Biases in Artificial Intelligence (AI) or Machine Learning (ML) systems due to skewed datasets problematise the application of prediction models in practice. Representation bias is a prevalent form of bias found in the majority of datasets. This bias arises when training data inadequately represents certain segments of the data space, resulting in poor generalisation of prediction models. Despite AI practitioners employing various methods to mitigate representation bias, their effectiveness is often limited due to a lack of thorough domain knowledge. To address this limitation, this paper introduces human-in-the-loop interaction approaches for representation debiasing of generated data involving domain experts. Our work advocates for a controlled data generation process involving domain experts to effectively mitigate the effects of representation bias. We argue that domain experts can leverage their expertise to assess how representation bias affects prediction models. Moreover, our interaction approaches can facilitate domain experts in steering data augmentation algorithms to produce debiased augmented data and validate or refine the generated samples to reduce representation bias. We also discuss how these approaches can be leveraged for designing and developing user-centred AI systems to mitigate the impact of representation bias through effective collaboration between domain experts and AI.
Title:
Uplifting Range-View-based 3D Semantic Segmentation in Real-Time with Multi-Sensor Fusion
Abstract
Range-View(RV)-based 3D point cloud segmentation is widely adopted due to its compact data form. However, RV-based methods fall short in providing robust segmentation for the occluded points and suffer from distortion of projected RGB images due to the sparse nature of 3D point clouds. To alleviate these problems, we propose a new LiDAR and Camera Range-view-based 3D point cloud semantic segmentation method (LaCRange). Specifically, a distortion-compensating knowledge distillation (DCKD) strategy is designed to remedy the adverse effect of RV projection of RGB images. Moreover, a context-based feature fusion module is introduced for robust and preservative sensor fusion. Finally, in order to address the limited resolution of RV and its insufficiency of 3D topology, a new point refinement scheme is devised for proper aggregation of features in 2D and augmentation of point features in 3D. We evaluated the proposed method on large-scale autonomous driving datasets \ie SemanticKITTI and nuScenes. In addition to being real-time, the proposed method achieves state-of-the-art results on nuScenes benchmark
Title:
TASeg: Temporal Aggregation Network for LiDAR Semantic Segmentation
Abstract
Training deep models for LiDAR semantic segmentation is challenging due to the inherent sparsity of point clouds. Utilizing temporal data is a natural remedy against the sparsity problem as it makes the input signal denser. However, previous multi-frame fusion algorithms fall short in utilizing sufficient temporal information due to the memory constraint, and they also ignore the informative temporal images. To fully exploit rich information hidden in long-term temporal point clouds and images, we present the Temporal Aggregation Network, termed TASeg. Specifically, we propose a Temporal LiDAR Aggregation and Distillation (TLAD) algorithm, which leverages historical priors to assign different aggregation steps for different classes. It can largely reduce memory and time overhead while achieving higher accuracy. Besides, TLAD trains a teacher injected with gt priors to distill the model, further boosting the performance. To make full use of temporal images, we design a Temporal Image Aggregation and Fusion (TIAF) module, which can greatly expand the camera FOV and enhance the present features. Temporal LiDAR points in the camera FOV are used as mediums to transform temporal image features to the present coordinate for temporal multi-modal fusion. Moreover, we develop a Static-Moving Switch Augmentation (SMSA) algorithm, which utilizes sufficient temporal information to enable objects to switch their motion states freely, thus greatly increasing static and moving training samples. Our TASeg ranks 1st on three challenging tracks, i.e., SemanticKITTI single-scan track, multi-scan track and nuScenes LiDAR segmentation track, strongly demonstrating the superiority of our method. Codes are available at this https URL.
Title:
Semi-supervised 3D Object Detection with PatchTeacher and PillarMix
Abstract
Semi-supervised learning aims to leverage numerous unlabeled data to improve the model performance. Current semi-supervised 3D object detection methods typically use a teacher to generate pseudo labels for a student, and the quality of the pseudo labels is essential for the final performance. In this paper, we propose PatchTeacher, which focuses on partial scene 3D object detection to provide high-quality pseudo labels for the student. Specifically, we divide a complete scene into a series of patches and feed them to our PatchTeacher sequentially. PatchTeacher leverages the low memory consumption advantage of partial scene detection to process point clouds with a high-resolution voxelization, which can minimize the information loss of quantization and extract more fine-grained features. However, it is non-trivial to train a detector on fractions of the scene. Therefore, we introduce three key techniques, i.e., Patch Normalizer, Quadrant Align, and Fovea Selection, to improve the performance of PatchTeacher. Moreover, we devise PillarMix, a strong data augmentation strategy that mixes truncated pillars from different LiDAR scans to generate diverse training samples and thus help the model learn more general representation. Extensive experiments conducted on Waymo and ONCE datasets verify the effectiveness and superiority of our method and we achieve new state-of-the-art results, surpassing existing methods by a large margin. Codes are available at this https URL.
Title:
Learning a Mini-batch Graph Transformer via Two-stage Interaction Augmentation
Abstract
Mini-batch Graph Transformer (MGT), as an emerging graph learning model, has demonstrated significant advantages in semi-supervised node prediction tasks with improved computational efficiency and enhanced model robustness. However, existing methods for processing local information either rely on sampling or simple aggregation, which respectively result in the loss and squashing of critical neighbor information.Moreover, the limited number of nodes in each mini-batch restricts the model's capacity to capture the global characteristic of the graph. In this paper, we propose LGMformer, a novel MGT model that employs a two-stage augmented interaction strategy, transitioning from local to global perspectives, to address the aforementioned bottlenecks.The local interaction augmentation (LIA) presents a neighbor-target interaction Transformer (NTIformer) to acquire an insightful understanding of the co-interaction patterns between neighbors and the target node, resulting in a locally effective token list that serves as input for the MGT. In contrast, global interaction augmentation (GIA) adopts a cross-attention mechanism to incorporate entire graph prototypes into the target node epresentation, thereby compensating for the global graph information to ensure a more comprehensive perception. To this end, LGMformer achieves the enhancement of node representations under the MGT paradigm.Experimental results related to node classification on the ten benchmark datasets demonstrate the effectiveness of the proposed method. Our code is available at this https URL.
Title:
Minimizing PLM-Based Few-Shot Intent Detectors
Authors: Haode Zhang, Xiao-Ming Wu, Albert Y.S. Lam
Subjects: Subjects:
Computation and Language (cs.CL)
Abstract
Recent research has demonstrated the feasibility of training efficient intent detectors based on pre-trained language model~(PLM) with limited labeled data. However, deploying these detectors in resource-constrained environments such as mobile devices poses challenges due to their large sizes. In this work, we aim to address this issue by exploring techniques to minimize the size of PLM-based intent detectors trained with few-shot data. Specifically, we utilize large language models (LLMs) for data augmentation, employ a cutting-edge model compression method for knowledge distillation, and devise a vocabulary pruning mechanism called V-Prune. Through these approaches, we successfully achieve a compression ratio of 21 in model memory usage, including both Transformer and the vocabulary, while maintaining almost identical performance levels on four real-world benchmarks.
Title:
Augmented Neural Fine-Tuning for Efficient Backdoor Purification
Authors: Nazmul Karim, Abdullah Al Arafat, Umar Khalid, Zhishan Guo, Nazanin Rahnavard
Subjects: Subjects:
Computer Vision and Pattern Recognition (cs.CV)
Abstract
Recent studies have revealed the vulnerability of deep neural networks (DNNs) to various backdoor attacks, where the behavior of DNNs can be compromised by utilizing certain types of triggers or poisoning mechanisms. State-of-the-art (SOTA) defenses employ too-sophisticated mechanisms that require either a computationally expensive adversarial search module for reverse-engineering the trigger distribution or an over-sensitive hyper-parameter selection module. Moreover, they offer sub-par performance in challenging scenarios, e.g., limited validation data and strong attacks. In this paper, we propose Neural mask Fine-Tuning (NFT) with an aim to optimally re-organize the neuron activities in a way that the effect of the backdoor is removed. Utilizing a simple data augmentation like MixUp, NFT relaxes the trigger synthesis process and eliminates the requirement of the adversarial search module. Our study further reveals that direct weight fine-tuning under limited validation data results in poor post-purification clean test accuracy, primarily due to overfitting issue. To overcome this, we propose to fine-tune neural masks instead of model weights. In addition, a mask regularizer has been devised to further mitigate the model drift during the purification process. The distinct characteristics of NFT render it highly efficient in both runtime and sample usage, as it can remove the backdoor even when a single sample is available from each class. We validate the effectiveness of NFT through extensive experiments covering the tasks of image classification, object detection, video action recognition, 3D point cloud, and natural language processing. We evaluate our method against 14 different attacks (LIRA, WaNet, etc.) on 11 benchmark data sets such as ImageNet, UCF101, Pascal VOC, ModelNet, OpenSubtitles2012, etc.
Title:
PARE-Net: Position-Aware Rotation-Equivariant Networks for Robust Point Cloud Registration
Authors: Runzhao Yao, Shaoyi Du, Wenting Cui, Canhui Tang, Chengwu Yang
Subjects: Subjects:
Computer Vision and Pattern Recognition (cs.CV)
Abstract
Learning rotation-invariant distinctive features is a fundamental requirement for point cloud registration. Existing methods often use rotation-sensitive networks to extract features, while employing rotation augmentation to learn an approximate invariant mapping rudely. This makes networks fragile to rotations, overweight, and hinders the distinctiveness of features. To tackle these problems, we propose a novel position-aware rotation-equivariant network, for efficient, light-weighted, and robust registration. The network can provide a strong model inductive bias to learn rotation-equivariant/invariant features, thus addressing the aforementioned limitations. To further improve the distinctiveness of descriptors, we propose a position-aware convolution, which can better learn spatial information of local structures. Moreover, we also propose a feature-based hypothesis proposer. It leverages rotation-equivariant features that encode fine-grained structure orientations to generate reliable model hypotheses. Each correspondence can generate a hypothesis, thus it is more efficient than classic estimators that require multiple reliable correspondences. Accordingly, a contrastive rotation loss is presented to enhance the robustness of rotation-equivariant features against data degradation. Extensive experiments on indoor and outdoor datasets demonstrate that our method significantly outperforms the SOTA methods in terms of registration recall while being lightweight and keeping a fast speed. Moreover, experiments on rotated datasets demonstrate its robustness against rotation variations. Code is available at this https URL.
Title:
The Hidden Influence of Latent Feature Magnitude When Learning with Imbalanced Data
Abstract
Machine learning (ML) models have difficulty generalizing when the number of training class instances are numerically imbalanced. The problem of generalization in the face of data imbalance has largely been attributed to the lack of training data for under-represented classes and to feature overlap. The typical remedy is to implement data augmentation for classes with fewer instances or to assign a higher cost to minority class prediction errors or to undersample the prevalent class. However, we show that one of the central causes of impaired generalization when learning with imbalanced data is the inherent manner in which ML models perform inference. These models have difficulty generalizing due to their heavy reliance on the magnitude of encoded signals. During inference, the models predict classes based on a combination of encoded signal magnitudes that linearly sum to the largest scalar. We demonstrate that even with aggressive data augmentation, which generally improves minority class prediction accuracy, parametric ML models still associate a class label with a limited number of feature combinations that sum to a prediction, which can affect generalization.
Title:
Few-Shot Bioacoustic Event Detection with Frame-Level Embedding Learning System
Authors: PengYuan Zhao, ChengWei Lu, Liang Zou
Subjects: Subjects:
Sound (cs.SD); Audio and Speech Processing (eess.AS)
Abstract
This technical report presents our frame-level embedding learning system for the DCASE2024 challenge for few-shot bioacoustic event detection (Task 5).In this work, we used log-mel and PCEN for feature extraction of the input audio, Netmamba Encoder as the information interaction network, and adopted data augmentation strategies to improve the generalizability of the trained model as well as multiple post-processing methods. Our final system achieved an F-measure score of 56.4%, securing the 2nd rank in the few-shot bioacoustic event detection category of the Detection and Classification of Acoustic Scenes and Events Challenge 2024.
Title:
Melon Fruit Detection and Quality Assessment Using Generative AI-Based Image Data Augmentation
Abstract
Monitoring and managing the growth and quality of fruits are very important tasks. To effectively train deep learning models like YOLO for real-time fruit detection, high-quality image datasets are essential. However, such datasets are often lacking in agriculture. Generative AI models can help create high-quality images. In this study, we used MidJourney and Firefly tools to generate images of melon greenhouses and post-harvest fruits through text-to-image, pre-harvest image-to-image, and post-harvest image-to-image methods. We evaluated these AIgenerated images using PSNR and SSIM metrics and tested the detection performance of the YOLOv9 model. We also assessed the net quality of real and generated fruits. Our results showed that generative AI could produce images very similar to real ones, especially for post-harvest fruits. The YOLOv9 model detected the generated images well, and the net quality was also measurable. This shows that generative AI can create realistic images useful for fruit detection and quality assessment, indicating its great potential in agriculture. This study highlights the potential of AI-generated images for data augmentation in melon fruit detection and quality assessment and envisions a positive future for generative AI applications in agriculture.
Title:
TCM-FTP: Fine-Tuning Large Language Models for Herbal Prescription Prediction
Authors: Xingzhi Zhou, Xin Dong, Chunhao Li, Yuning Bai, Yulong Xu, Ka Chun Cheung, Simon See, Xinpeng Song, Runshun Zhang, Xuezhong Zhou, Nevin L. Zhang
Subjects: Subjects:
Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Computational Engineering, Finance, and Science (cs.CE)
Abstract
Traditional Chinese medicine (TCM) relies on specific combinations of herbs in prescriptions to treat symptoms and signs, a practice that spans thousands of years. Predicting TCM prescriptions presents a fascinating technical challenge with practical implications. However, this task faces limitations due to the scarcity of high-quality clinical datasets and the intricate relationship between symptoms and herbs. To address these issues, we introduce DigestDS, a new dataset containing practical medical records from experienced experts in digestive system diseases. We also propose a method, TCM-FTP (TCM Fine-Tuning Pre-trained), to leverage pre-trained large language models (LLMs) through supervised fine-tuning on DigestDS. Additionally, we enhance computational efficiency using a low-rank adaptation technique. TCM-FTP also incorporates data augmentation by permuting herbs within prescriptions, capitalizing on their order-agnostic properties. Impressively, TCM-FTP achieves an F1-score of 0.8031, surpassing previous methods significantly. Furthermore, it demonstrates remarkable accuracy in dosage prediction, achieving a normalized mean square error of 0.0604. In contrast, LLMs without fine-tuning perform poorly. Although LLMs have shown capabilities on a wide range of tasks, this work illustrates the importance of fine-tuning for TCM prescription prediction, and we have proposed an effective way to do that.
Title:
An evaluation of CNN models and data augmentation techniques in hierarchical localization of mobile robots
Authors: J.J. Cabrera, O. J. Céspedes, S. Cebollada, O. Reinoso, L. Payá
Abstract
This work presents an evaluation of CNN models and data augmentation to carry out the hierarchical localization of a mobile robot by using omnidireccional images. In this sense, an ablation study of different state-of-the-art CNN models used as backbone is presented and a variety of data augmentation visual effects are proposed for addressing the visual localization of the robot. The proposed method is based on the adaption and re-training of a CNN with a dual purpose: (1) to perform a rough localization step in which the model is used to predict the room from which an image was captured, and (2) to address the fine localization step, which consists in retrieving the most similar image of the visual map among those contained in the previously predicted room by means of a pairwise comparison between descriptors obtained from an intermediate layer of the CNN. In this sense, we evaluate the impact of different state-of-the-art CNN models such as ConvNeXt for addressing the proposed localization. Finally, a variety of data augmentation visual effects are separately employed for training the model and their impact is assessed. The performance of the resulting CNNs is evaluated under real operation conditions, including changes in the lighting conditions. Our code is publicly available on the project website this https URL
Title:
Mitigating Data Imbalance for Software Vulnerability Assessment: Does Data Augmentation Help?
Abstract
Background: Software Vulnerability (SV) assessment is increasingly adopted to address the ever-increasing volume and complexity of SVs. Data-driven approaches have been widely used to automate SV assessment tasks, particularly the prediction of the Common Vulnerability Scoring System (CVSS) metrics such as exploitability, impact, and severity. SV assessment suffers from the imbalanced distributions of the CVSS classes, but such data imbalance has been hardly understood and addressed in the literature. Aims: We conduct a large-scale study to quantify the impacts of data imbalance and mitigate the issue for SV assessment through the use of data augmentation. Method: We leverage nine data augmentation techniques to balance the class distributions of the CVSS metrics. We then compare the performance of SV assessment models with and without leveraging the augmented data. Results: Through extensive experiments on 180k+ real-world SVs, we show that mitigating data imbalance can significantly improve the predictive performance of models for all the CVSS tasks, by up to 31.8% in Matthews Correlation Coefficient. We also discover that simple text augmentation like combining random text insertion, deletion, and replacement can outperform the baseline across the board. Conclusions: Our study provides the motivation and the first promising step toward tackling data imbalance for effective SV assessment.
Title:
Domain Generalization for 6D Pose Estimation Through NeRF-based Image Synthesis
Authors: Antoine Legrand, Renaud Detry, Christophe De Vleeschouwer
Subjects: Subjects:
Computer Vision and Pattern Recognition (cs.CV); Image and Video Processing (eess.IV)
Abstract
This work introduces a novel augmentation method that increases the diversity of a train set to improve the generalization abilities of a 6D pose estimation network. For this purpose, a Neural Radiance Field is trained from synthetic images and exploited to generate an augmented set. Our method enriches the initial set by enabling the synthesis of images with (i) unseen viewpoints, (ii) rich illumination conditions through appearance extrapolation, and (iii) randomized textures. We validate our augmentation method on the challenging use-case of spacecraft pose estimation and show that it significantly improves the pose estimation generalization capabilities. On the SPEED+ dataset, our method reduces the error on the pose by 50% on both target domains.
Title:
Leveraging LLM-Respondents for Item Evaluation: a Psychometric Analysis
Authors: Yunting Liu, Shreya Bhandari, Zachary A. Pardos
Subjects: Subjects:
Computers and Society (cs.CY); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
Abstract
Effective educational measurement relies heavily on the curation of well-designed item pools (i.e., possessing the right psychometric properties). However, item calibration is time-consuming and costly, requiring a sufficient number of respondents for the response process. We explore using six different LLMs (GPT-3.5, GPT-4, Llama 2, Llama 3, Gemini-Pro, and Cohere Command R Plus) and various combinations of them using sampling methods to produce responses with psychometric properties similar to human answers. Results show that some LLMs have comparable or higher proficiency in College Algebra than college students. No single LLM mimics human respondents due to narrow proficiency distributions, but an ensemble of LLMs can better resemble college students' ability distribution. The item parameters calibrated by LLM-Respondents have high correlations (e.g. > 0.8 for GPT-3.5) compared to their human calibrated counterparts, and closely resemble the parameters of the human subset (e.g. 0.02 Spearman correlation difference). Several augmentation strategies are evaluated for their relative performance, with resampling methods proving most effective, enhancing the Spearman correlation from 0.89 (human only) to 0.93 (augmented human).
Title:
STARS: Self-supervised Tuning for 3D Action Recognition in Skeleton Sequences
Authors: Soroush Mehraban, Mohammad Javad Rajabi, Babak Taati
Subjects: Subjects:
Computer Vision and Pattern Recognition (cs.CV)
Abstract
Self-supervised pretraining methods with masked prediction demonstrate remarkable within-dataset performance in skeleton-based action recognition. However, we show that, unlike contrastive learning approaches, they do not produce well-separated clusters. Additionally, these methods struggle with generalization in few-shot settings. To address these issues, we propose Self-supervised Tuning for 3D Action Recognition in Skeleton sequences (STARS). Specifically, STARS first uses a masked prediction stage using an encoder-decoder architecture. It then employs nearest-neighbor contrastive learning to partially tune the weights of the encoder, enhancing the formation of semantic clusters for different actions. By tuning the encoder for a few epochs, and without using hand-crafted data augmentations, STARS achieves state-of-the-art self-supervised results in various benchmarks, including NTU-60, NTU-120, and PKU-MMD. In addition, STARS exhibits significantly better results than masked prediction models in few-shot settings, where the model has not seen the actions throughout pretraining. Project page: this https URL
Keyword: detection
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Keyword: face recognition
Title:
Title:
Keyword: augmentation
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title: