Detecting Two Objects with the Same Center

Search before asking

[X] I have searched the YOLOv5 issues and discussions and found no similar questions.

Question

Hello,

I have encountered a problem while working with yolov5 on my custom data. The problem is - If there are two objects that have the same center point, the model fails to detect the inner object. The model detects both objects with low confidence and in the exact same shape. I found that increasing the imgsz might help but still, the results are pretty bad. I wanted to ask if there is a known solution to this problem or any ideas you might have since I didn't find any solutions. I am attaching two images, one with the original bounding boxes, and the other with the predicted bounding boxes. In these images I am trying to detect the eye and the iris, and their bounding boxes share almost the exact same center point.

Thank you!

nadav nadav

Additional

No response

👋 Hello @nadavkirsch-6over6, thank you for your interest in YOLOv5 🚀! Please visit our ⭐️ Tutorials to get started, where you can find quickstart guides for simple tasks like Custom Data Training all the way to advanced concepts like Hyperparameter Evolution.

If this is a 🐛 Bug Report, please provide screenshots and minimum viable code to reproduce your issue, otherwise we can not help you.

If this is a custom training ❓ Question, please provide as much information as possible, including dataset images, training logs, screenshots, and a public link to online W&B logging if available.

For business inquiries or professional support requests please visit https://ultralytics.com or email support@ultralytics.com.

Requirements

Python>=3.7.0 with all requirements.txt installed including PyTorch>=1.7. To get started:

git clone https://github.com/ultralytics/yolov5  # clone
cd yolov5
pip install -r requirements.txt  # install

Environments

YOLOv5 may be run in any of the following up-to-date verified environments (with all dependencies including CUDA/CUDNN, Python and PyTorch preinstalled):

Google Colab and Kaggle notebooks with free GPU:
Google Cloud Deep Learning VM. See GCP Quickstart Guide
Amazon Deep Learning AMI. See AWS Quickstart Guide
Docker Image. See Docker Quickstart Guide

Status

If this badge is green, all YOLOv5 GitHub Actions Continuous Integration (CI) tests are currently passing. CI tests verify correct operation of YOLOv5 training (train.py), validation (val.py), inference (detect.py) and export (export.py) on macOS, Windows, and Ubuntu every 24 hours and on every commit.

@nadavkirsch-6over6 👋 Hello! Thanks for asking about improving YOLOv5 🚀 training results.

Identical centerpoints may result in ambiguity about which box to predict. Some architectures like FCOS attempt to solve this by configuring the prediction space differently, but before considering any exotic solutions I would review the recommendations below to verify your dataset is sufficiently aligned with best practices.

Most of the time good results can be obtained with no changes to the models or training settings, provided your dataset is sufficiently large and well labelled. If at first you don't get good results, there are steps you might be able to take to improve, but we always recommend users first train with all default settings before considering any changes. This helps establish a performance baseline and spot areas for improvement.

If you have questions about your training results we recommend you provide the maximum amount of information possible if you expect a helpful response, including results plots (train losses, val losses, P, R, mAP), PR curve, confusion matrix, training mosaics, test results and dataset statistics images such as labels.png. All of these are located in your project/name directory, typically yolov5/runs/train/exp.

We've put together a full guide for users looking to get the best results on their YOLOv5 trainings below.

Dataset

Images per class. ≥ 1500 images per class recommended
Instances per class. ≥ 10000 instances (labeled objects) per class recommended
Image variety. Must be representative of deployed environment. For real-world use cases we recommend images from different times of day, different seasons, different weather, different lighting, different angles, different sources (scraped online, collected locally, different cameras) etc.
Label consistency. All instances of all classes in all images must be labelled. Partial labelling will not work.
Label accuracy. Labels must closely enclose each object. No space should exist between an object and it's bounding box. No objects should be missing a label.
Label verification. View train_batch*.jpg on train start to verify your labels appear correct, i.e. see example mosaic.
Background images. Background images are images with no objects that are added to a dataset to reduce False Positives (FP). We recommend about 0-10% background images to help reduce FPs (COCO has 1000 background images for reference, 1% of the total). No labels are required for background images.

Model Selection

Larger models like YOLOv5x and YOLOv5x6 will produce better results in nearly all cases, but have more parameters, require more CUDA memory to train, and are slower to run. For mobile deployments we recommend YOLOv5s/m, for cloud deployments we recommend YOLOv5l/x. See our README table for a full comparison of all models.

YOLOv5 Models

Start from Pretrained weights. Recommended for small to medium sized datasets (i.e. VOC, VisDrone, GlobalWheat). Pass the name of the model to the --weights argument. Models download automatically from the latest YOLOv5 release.

python train.py --data custom.yaml --weights yolov5s.pt
                                         yolov5m.pt
                                         yolov5l.pt
                                         yolov5x.pt
                                         custom_pretrained.pt

Start from Scratch. Recommended for large datasets (i.e. COCO, Objects365, OIv6). Pass the model architecture yaml you are interested in, along with an empty --weights '' argument:

python train.py --data custom.yaml --weights '' --cfg yolov5s.yaml
                                                  yolov5m.yaml
                                                  yolov5l.yaml
                                                  yolov5x.yaml

Training Settings

Before modifying anything, first train with default settings to establish a performance baseline. A full list of train.py settings can be found in the train.py argparser.

Epochs. Start with 300 epochs. If this overfits early then you can reduce epochs. If overfitting does not occur after 300 epochs, train longer, i.e. 600, 1200 etc epochs.
Image size. COCO trains at native resolution of --img 640, though due to the high amount of small objects in the dataset it can benefit from training at higher resolutions such as --img 1280. If there are many small objects then custom datasets will benefit from training at native or higher resolution. Best inference results are obtained at the same --img as the training was run at, i.e. if you train at --img 1280 you should also test and detect at --img 1280.
Batch size. Use the largest --batch-size that your hardware allows for. Small batch sizes produce poor batchnorm statistics and should be avoided.
Hyperparameters. Default hyperparameters are in hyp.scratch-low.yaml. We recommend you train with default hyperparameters first before thinking of modifying any. In general, increasing augmentation hyperparameters will reduce and delay overfitting, allowing for longer trainings and higher final mAP. Reduction in loss component gain hyperparameters like hyp['obj'] will help reduce overfitting in those specific loss components. For an automated method of optimizing these hyperparameters, see our Hyperparameter Evolution Tutorial.

Alternatively, since you appear to be detecting an eye in addition to a specific part of the eye, perhaps you could detect the eye, cut out the portion of the image in the detection, and then run your second stage detector on just the image of the eye? It would be for more expensive computationally, but much simpler to see what is really happening. Additionally it would be easy to collect a lot of source images of just eyes to build your iris dataset. I am doing something similar to do different classifications on detected objects and am happy with the initial results

👋 Hello, this issue has been automatically marked as stale because it has not had recent activity. Please note it will be closed if no further activity occurs.

Access additional YOLOv5 🚀 resources:

Wiki – https://github.com/ultralytics/yolov5/wiki
Tutorials – https://docs.ultralytics.com/yolov5
Docs – https://docs.ultralytics.com

Access additional Ultralytics ⚡ resources:

Ultralytics HUB – https://ultralytics.com/hub
Vision API – https://ultralytics.com/yolov5
About Us – https://ultralytics.com/about
Join Our Team – https://ultralytics.com/work
Contact Us – https://ultralytics.com/contact

Feel free to inform us of any other issues you discover or feature requests that come to mind in the future. Pull Requests (PRs) are also always welcomed!

Thank you for your contributions to YOLOv5 🚀 and Vision AI ⭐!

@gillonba great questions! 🤔

The "only look once" nature of YOLO refers to its approach of predicting all bounding boxes and class probabilities from a single forward pass of the network, combining high levels of precision with real-time speeds. While running two models and combining results can work, it would indeed be more computationally expensive and time-consuming compared to a single, optimized model.

Your alternative approach of first detecting the entire eye and then running a second stage detector on just the eye image is a promising strategy! This two-stage detection method could potentially simplify the detection process and even improve the overall accuracy, while also allowing for easier dataset collection as you mentioned.

Your results sound impressive! It's great to hear that you are happy with the initial outcomes. Keep up the great work and don't hesitate to reach out if you need further assistance!

ultralytics / yolov5