New submissions for Fri, 11 Mar 22

[zhuhu00]

Keyword: SLAM

SelfTune: Metrically Scaled Monocular Depth Estimation through Self-Supervised Learning

• Authors: Jaehoon Choi, Dongki Jung, Yonghan Lee, Deokhwa Kim, Dinesh Manocha, Donghwan Lee
• Subjects: Computer Vision and Pattern Recognition (cs.CV)
• Arxiv link: https://arxiv.org/abs/2203.05332
• Pdf link: https://arxiv.org/pdf/2203.05332
• Abstract Monocular depth estimation in the wild inherently predicts depth up to an unknown scale. To resolve scale ambiguity issue, we present a learning algorithm that leverages monocular simultaneous localization and mapping (SLAM) with proprioceptive sensors. Such monocular SLAM systems can provide metrically scaled camera poses. Given these metric poses and monocular sequences, we propose a self-supervised learning method for the pre-trained supervised monocular depth networks to enable metrically scaled depth estimation. Our approach is based on a teacher-student formulation which guides our network to predict high-quality depths. We demonstrate that our approach is useful for various applications such as mobile robot navigation and is applicable to diverse environments. Our full system shows improvements over recent self-supervised depth estimation and completion methods on EuRoC, OpenLORIS, and ScanNet datasets.

  Keyword: Visual inertial

  There is no result

  Keyword: livox

  There is no result

  Keyword: loam

  There is no result

  Keyword: Visual inertial odometry

  There is no result

  Keyword: lidar

  Crowd Source Scene Change Detection and Local Map Update

• Authors: Itzik Wilf, Nati Daniel, Lin Manqing, Firas Shama, Omri Asraf, Feng Wensen, Ofer Kruzel
• Subjects: Computer Vision and Pattern Recognition (cs.CV)
• Arxiv link: https://arxiv.org/abs/2203.05205
• Pdf link: https://arxiv.org/pdf/2203.05205
• Abstract As scene changes with time map descriptors become outdated, affecting VPS localization accuracy. In this work, we propose an approach to detect structural and texture scene changes to be followed by map update. In our method - map includes 3D points with descriptors generated either via LiDAR or SFM. Common approaches suffer from shortcomings: 1) Direct comparison of the two point-clouds for change detection is slow due to the need to build new point-cloud every time we want to compare; 2) Image based comparison requires to keep the map images adding substantial storage overhead. To circumvent this problems, we propose an approach based on point-clouds descriptors comparison: 1) Based on VPS poses select close query and map images pairs, 2) Registration of query images to map image descriptors, 3) Use segmentation to filter out dynamic or short term temporal changes, 4) Compare the descriptors between corresponding segments.

  Keyword: loop detection

  There is no result

  Keyword: autonomous driving

  Adaptive Trajectory Prediction via Transferable GNN

• Authors: Yi Xu, Lichen Wang, Yizhou Wang, Yun Fu
• Subjects: Computer Vision and Pattern Recognition (cs.CV)
• Arxiv link: https://arxiv.org/abs/2203.05046
• Pdf link: https://arxiv.org/pdf/2203.05046
• Abstract Pedestrian trajectory prediction is an essential component in a wide range of AI applications such as autonomous driving and robotics. Existing methods usually assume the training and testing motions follow the same pattern while ignoring the potential distribution differences (e.g., shopping mall and street). This issue results in inevitable performance decrease. To address this issue, we propose a novel Transferable Graph Neural Network (T-GNN) framework, which jointly conducts trajectory prediction as well as domain alignment in a unified framework. Specifically, a domain invariant GNN is proposed to explore the structural motion knowledge where the domain specific knowledge is reduced. Moreover, an attention-based adaptive knowledge learning module is further proposed to explore fine-grained individual-level feature representation for knowledge transfer. By this way, disparities across different trajectory domains will be better alleviated. More challenging while practical trajectory prediction experiments are designed, and the experimental results verify the superior performance of our proposed model. To the best of our knowledge, our work is the pioneer which fills the gap in benchmarks and techniques for practical pedestrian trajectory prediction across different domains.

  Regret-Matching Learning-Based Task Assignment in Vehicular Edge Computing

• Authors: Bach Long Nguyen, Duong D. Nguyen, Hung X. Nguyen, Duy T. Ngo
• Subjects: Systems and Control (eess.SY)
• Arxiv link: https://arxiv.org/abs/2203.05281
• Pdf link: https://arxiv.org/pdf/2203.05281
• Abstract Vehicular edge computing has recently been proposed to support computation-intensive applications in Intelligent Transportation Systems (ITS) such as augmented reality and autonomous driving. Despite recent progress in this area, significant challenges remain to efficiently allocate limited computation resources to a range of time-critical ITS tasks. Toward this end, the current paper develops a new task assignment scheme for vehicles in a highway scenario. Because of the high speed of vehicles and the limited communication range of road side units (RSUs), the computation tasks of participating vehicles are to be migrated across multiple servers. We formulate a binary nonlinear programming (BNLP) problem of assigning computation tasks from vehicles to RSUs and a macrocell base station. To deal with the potentially large size of the formulated optimization problem, we develop a distributed multi-agent regret-matching learning algorithm. Based on the regret minimization principle, the proposed algorithm employs a forgetting method that allows the learning process to quickly adapt to and effectively handle the high mobility feature of vehicle networks. We theoretically prove that it converges to the correlated equilibrium solutions of the considered BNLP problem. Simulation results with practical parameter settings show that the proposed algorithm offers the lowest total delay and cost of processing tasks. Importantly, our algorithm converges much faster than existing methods as the problem size grows, demonstrating its clear advantage in large-scale vehicular networks.

  SoK: On the Semantic AI Security in Autonomous Driving

• Authors: Junjie Shen, Ningfei Wang, Ziwen Wan, Yunpeng Luo, Takami Sato, Zhisheng Hu, Xinyang Zhang, Shengjian Guo, Zhenyu Zhong, Kang Li, Ziming Zhao, Chunming Qiao, Qi Alfred Chen
• Subjects: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Robotics (cs.RO)
• Arxiv link: https://arxiv.org/abs/2203.05314
• Pdf link: https://arxiv.org/pdf/2203.05314
• Abstract Autonomous Driving (AD) systems rely on AI components to make safety and correct driving decisions. Unfortunately, today's AI algorithms are known to be generally vulnerable to adversarial attacks. However, for such AI component-level vulnerabilities to be semantically impactful at the system level, it needs to address non-trivial semantic gaps both (1) from the system-level attack input spaces to those at AI component level, and (2) from AI component-level attack impacts to those at the system level. In this paper, we define such research space as semantic AI security as opposed to generic AI security. Over the past 5 years, increasingly more research works are performed to tackle such semantic AI security challenges in AD context, which has started to show an exponential growth trend. In this paper, we perform the first systematization of knowledge of such growing semantic AD AI security research space. In total, we collect and analyze 53 such papers, and systematically taxonomize them based on research aspects critical for the security field. We summarize 6 most substantial scientific gaps observed based on quantitative comparisons both vertically among existing AD AI security works and horizontally with security works from closely-related domains. With these, we are able to provide insights and potential future directions not only at the design level, but also at the research goal, methodology, and community levels. To address the most critical scientific methodology-level gap, we take the initiative to develop an open-source, uniform, and extensible system-driven evaluation platform, named PASS, for the semantic AD AI security research community. We also use our implemented platform prototype to showcase the capabilities and benefits of such a platform using representative semantic AD AI attacks.

  Keyword: mapping

  On the influence of over-parameterization in manifold based surrogates and deep neural operators

• Authors: Katiana Kontolati, Somdatta Goswami, Michael D. Shields, George Em Karniadakis
• Subjects: Machine Learning (cs.LG)
• Arxiv link: https://arxiv.org/abs/2203.05071
• Pdf link: https://arxiv.org/pdf/2203.05071
• Abstract Constructing accurate and generalizable approximators for complex physico-chemical processes exhibiting highly non-smooth dynamics is challenging. In this work, we propose new developments and perform comparisons for two promising approaches: manifold-based polynomial chaos expansion (m-PCE) and the deep neural operator (DeepONet), and we examine the effect of over-parameterization on generalization. We demonstrate the performance of these methods in terms of generalization accuracy by solving the 2D time-dependent Brusselator reaction-diffusion system with uncertainty sources, modeling an autocatalytic chemical reaction between two species. We first propose an extension of the m-PCE by constructing a mapping between latent spaces formed by two separate embeddings of input functions and output QoIs. To enhance the accuracy of the DeepONet, we introduce weight self-adaptivity in the loss function. We demonstrate that the performance of m-PCE and DeepONet is comparable for cases of relatively smooth input-output mappings. However, when highly non-smooth dynamics is considered, DeepONet shows higher accuracy. We also find that for m-PCE, modest over-parameterization leads to better generalization, both within and outside of distribution, whereas aggressive over-parameterization leads to over-fitting. In contrast, an even highly over-parameterized DeepONet leads to better generalization for both smooth and non-smooth dynamics. Furthermore, we compare the performance of the above models with another operator learning model, the Fourier Neural Operator, and show that its over-parameterization also leads to better generalization. Our studies show that m-PCE can provide very good accuracy at very low training cost, whereas a highly over-parameterized DeepONet can provide better accuracy and robustness to noise but at higher training cost. In both methods, the inference cost is negligible.

  MetAug: Contrastive Learning via Meta Feature Augmentation

• Authors: Jiangmeng Li, Wenwen Qiang, Changwen Zheng, Bing Su, Hui Xiong
• Subjects: Computer Vision and Pattern Recognition (cs.CV)
• Arxiv link: https://arxiv.org/abs/2203.05119
• Pdf link: https://arxiv.org/pdf/2203.05119
• Abstract What matters for contrastive learning? We argue that contrastive learning heavily relies on informative features, or "hard" (positive or negative) features. Early works include more informative features by applying complex data augmentations and large batch size or memory bank, and recent works design elaborate sampling approaches to explore informative features. The key challenge toward exploring such features is that the source multi-view data is generated by applying random data augmentations, making it infeasible to always add useful information in the augmented data. Consequently, the informativeness of features learned from such augmented data is limited. In response, we propose to directly augment the features in latent space, thereby learning discriminative representations without a large amount of input data. We perform a meta learning technique to build the augmentation generator that updates its network parameters by considering the performance of the encoder. However, insufficient input data may lead the encoder to learn collapsed features and therefore malfunction the augmentation generator. A new margin-injected regularization is further added in the objective function to avoid the encoder learning a degenerate mapping. To contrast all features in one gradient back-propagation step, we adopt the proposed optimization-driven unified contrastive loss instead of the conventional contrastive loss. Empirically, our method achieves state-of-the-art results on several benchmark datasets.

  Stable Parametrization of Continuous and Piecewise-Linear Functions

• Authors: Alexis Goujon, Joaquim Campos, Michael Unser
• Subjects: Numerical Analysis (math.NA)
• Arxiv link: https://arxiv.org/abs/2203.05261
• Pdf link: https://arxiv.org/pdf/2203.05261
• Abstract Rectified-linear-unit (ReLU) neural networks, which play a prominent role in deep learning, generate continuous and piecewise-linear (CPWL) functions. While they provide a powerful parametric representation, the mapping between the parameter and function spaces lacks stability. In this paper, we investigate an alternative representation of CPWL functions that relies on local hat basis functions. It is predicated on the fact that any CPWL function can be specified by a triangulation and its values at the grid points. We give the necessary and sufficient condition on the triangulation (in any number of dimensions) for the hat functions to form a Riesz basis, which ensures that the link between the parameters and the corresponding CPWL function is stable and unique. In addition, we provide an estimate of the $\ell_2\rightarrow L_2$ condition number of this local representation. Finally, as a special case of our framework, we focus on a systematic parametrization of $\mathbb{R}^d$ with control points placed on a uniform grid. In particular, we choose hat basis functions that are shifted replicas of a single linear box spline. In this setting, we prove that our general estimate of the condition number is optimal. We also relate our local representation to a nonlocal one based on shifts of a causal ReLU-like function.

  API: Boosting Multi-Agent Reinforcement Learning via Agent-Permutation-Invariant Networks

• Authors: Xiaotian Hao, Weixun Wang, Hangyu Mao, Yaodong Yang, Dong Li, Yan Zheng, Zhen Wang, Jianye Hao
• Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA)
• Arxiv link: https://arxiv.org/abs/2203.05285
• Pdf link: https://arxiv.org/pdf/2203.05285
• Abstract Multi-agent reinforcement learning suffers from poor sample efficiency due to the exponential growth of the state-action space. Considering a homogeneous multiagent system, a global state consisting of $m$ homogeneous components has $m!$ differently ordered representations, thus designing functions satisfying permutation invariant (PI) can reduce the state space by a factor of $\frac{1}{m!}$. However, mainstream MARL algorithms ignore this property and learn over the original state space. To achieve PI, previous works including data augmentation based methods and embedding-sharing architecture based methods, suffer from training instability and limited model capacity. In this work, we propose two novel designs to achieve PI, while avoiding the above limitations. The first design permutes the same but differently ordered inputs back to the same order and the downstream networks only need to learn function mapping over fixed-ordering inputs instead of all permutations, which is much easier to train. The second design applies a hypernetwork to generate customized embedding for each component, which has higher representational capacity than the previous embedding-sharing method. Empirical results on the SMAC benchmark show that the proposed method achieves 100% win-rates in almost all hard and super-hard scenarios (never achieved before), and superior sample-efficiency than the state-of-the-art baselines by up to 400%.

  SelfTune: Metrically Scaled Monocular Depth Estimation through Self-Supervised Learning

• Authors: Jaehoon Choi, Dongki Jung, Yonghan Lee, Deokhwa Kim, Dinesh Manocha, Donghwan Lee
• Subjects: Computer Vision and Pattern Recognition (cs.CV)
• Arxiv link: https://arxiv.org/abs/2203.05332
• Pdf link: https://arxiv.org/pdf/2203.05332
• Abstract Monocular depth estimation in the wild inherently predicts depth up to an unknown scale. To resolve scale ambiguity issue, we present a learning algorithm that leverages monocular simultaneous localization and mapping (SLAM) with proprioceptive sensors. Such monocular SLAM systems can provide metrically scaled camera poses. Given these metric poses and monocular sequences, we propose a self-supervised learning method for the pre-trained supervised monocular depth networks to enable metrically scaled depth estimation. Our approach is based on a teacher-student formulation which guides our network to predict high-quality depths. We demonstrate that our approach is useful for various applications such as mobile robot navigation and is applicable to diverse environments. Our full system shows improvements over recent self-supervised depth estimation and completion methods on EuRoC, OpenLORIS, and ScanNet datasets.

  Keyword: localization

  Connecting sufficient conditions for domain adaptation: source-guided uncertainty, relaxed divergences and discrepancy localization

• Authors: Sofien Dhouib, Setareh Maghsudi
• Subjects: Machine Learning (cs.LG)
• Arxiv link: https://arxiv.org/abs/2203.05076
• Pdf link: https://arxiv.org/pdf/2203.05076
• Abstract Recent advances in domain adaptation establish that requiring a low risk on the source domain and equal feature marginals degrade the adaptation's performance. At the same time, empirical evidence shows that incorporating an unsupervised target domain term that pushes decision boundaries away from the high-density regions, along with relaxed alignment, improves adaptation. In this paper, we theoretically justify such observations via a new bound on the target risk, and we connect two notions of relaxation for divergence, namely $\beta-$relaxed divergences and localization. This connection allows us to incorporate the source domain's categorical structure into the relaxation of the considered divergence, provably resulting in a better handling of the label shift case in particular.

  OpenTAL: Towards Open Set Temporal Action Localization

• Authors: Wentao Bao, Qi Yu, Yu Kong
• Subjects: Computer Vision and Pattern Recognition (cs.CV)
• Arxiv link: https://arxiv.org/abs/2203.05114
• Pdf link: https://arxiv.org/pdf/2203.05114
• Abstract Temporal Action Localization (TAL) has experienced remarkable success under the supervised learning paradigm. However, existing TAL methods are rooted in the closed set assumption, which cannot handle the inevitable unknown actions in open-world scenarios. In this paper, we, for the first time, step toward the Open Set TAL (OSTAL) problem and propose a general framework OpenTAL based on Evidential Deep Learning (EDL). Specifically, the OpenTAL consists of uncertainty-aware action classification, actionness prediction, and temporal location regression. With the proposed importance-balanced EDL method, classification uncertainty is learned by collecting categorical evidence majorly from important samples. To distinguish the unknown actions from background video frames, the actionness is learned by the positive-unlabeled learning. The classification uncertainty is further calibrated by leveraging the guidance from the temporal localization quality. The OpenTAL is general to enable existing TAL models for open set scenarios, and experimental results on THUMOS14 and ActivityNet1.3 benchmarks show the effectiveness of our method. The code and pre-trained models are released at https://www.rit.edu/actionlab/opental.

  Crowd Source Scene Change Detection and Local Map Update

• Authors: Itzik Wilf, Nati Daniel, Lin Manqing, Firas Shama, Omri Asraf, Feng Wensen, Ofer Kruzel
• Subjects: Computer Vision and Pattern Recognition (cs.CV)
• Arxiv link: https://arxiv.org/abs/2203.05205
• Pdf link: https://arxiv.org/pdf/2203.05205
• Abstract As scene changes with time map descriptors become outdated, affecting VPS localization accuracy. In this work, we propose an approach to detect structural and texture scene changes to be followed by map update. In our method - map includes 3D points with descriptors generated either via LiDAR or SFM. Common approaches suffer from shortcomings: 1) Direct comparison of the two point-clouds for change detection is slow due to the need to build new point-cloud every time we want to compare; 2) Image based comparison requires to keep the map images adding substantial storage overhead. To circumvent this problems, we propose an approach based on point-clouds descriptors comparison: 1) Based on VPS poses select close query and map images pairs, 2) Registration of query images to map image descriptors, 3) Use segmentation to filter out dynamic or short term temporal changes, 4) Compare the descriptors between corresponding segments.

  SelfTune: Metrically Scaled Monocular Depth Estimation through Self-Supervised Learning

• Authors: Jaehoon Choi, Dongki Jung, Yonghan Lee, Deokhwa Kim, Dinesh Manocha, Donghwan Lee
• Subjects: Computer Vision and Pattern Recognition (cs.CV)
• Arxiv link: https://arxiv.org/abs/2203.05332
• Pdf link: https://arxiv.org/pdf/2203.05332
• Abstract Monocular depth estimation in the wild inherently predicts depth up to an unknown scale. To resolve scale ambiguity issue, we present a learning algorithm that leverages monocular simultaneous localization and mapping (SLAM) with proprioceptive sensors. Such monocular SLAM systems can provide metrically scaled camera poses. Given these metric poses and monocular sequences, we propose a self-supervised learning method for the pre-trained supervised monocular depth networks to enable metrically scaled depth estimation. Our approach is based on a teacher-student formulation which guides our network to predict high-quality depths. We demonstrate that our approach is useful for various applications such as mobile robot navigation and is applicable to diverse environments. Our full system shows improvements over recent self-supervised depth estimation and completion methods on EuRoC, OpenLORIS, and ScanNet datasets.

  EyeLoveGAN: Exploiting domain-shifts to boost network learning with cycleGANs

• Authors: Josefine Vilsbøll Sundgaard, Kristine Aavild Juhl, Jakob Mølkjær Slipsager
• Subjects: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
• Arxiv link: https://arxiv.org/abs/2203.05344
• Pdf link: https://arxiv.org/pdf/2203.05344
• Abstract This paper presents our contribution to the REFUGE challenge 2020. The challenge consisted of three tasks based on a dataset of retinal images: Segmentation of optic disc and cup, classification of glaucoma, and localization of fovea. We propose employing convolutional neural networks for all three tasks. Segmentation is performed using a U-Net, classification is performed by a pre-trained InceptionV3 network, and fovea detection is performed by employing stacked hour-glass for heatmap prediction. The challenge dataset contains images from three different data sources. To enhance performance, cycleGANs were utilized to create a domain-shift between the data sources. These cycleGANs move images across domains, thus creating artificial images which can be used for training.