Segmentation Prediction: Saved Text File Normalisation

[Bampson]

Search before asking

• I have searched the YOLOv5 issues and discussions and found no similar questions.

Question

Maybe a bug? Noticing that the txt file generated after segmentation prediction is outputting different values to the object detection detection:

eg.
3 780 588 778 590 772 590 772 593 768 598 765 598 762 600 762 603 740 ... 0.464866

Should these be normalised under 1?

Additional

using: --save-conf --save-txt

[github-actions]

👋 Hello @Bampson, 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):

• 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, validation, inference, export and benchmarks on MacOS, Windows, and Ubuntu every 24 hours and on every commit.

[WongVi]

@Bampson
have you trained the segmentation network with a custom dataset?
could you please let me know how can you prepare data?

[glenn-jocher]

@Bampson yes these should be normalized. If they are not then there is a bug.

[glenn-jocher]

@Bampson can reproduce bug. Have added bug and TODO labels to this.

[glenn-jocher]

@Bampson good news 😃! Your original issue may now be fixed ✅ in PR #10205. Segmentation predictions are now outputted in normalized coordinates just like detection predictions.

To receive this update:

• Git – git pull from within your yolov5/ directory or git clone https://github.com/ultralytics/yolov5 again
• PyTorch Hub – Force-reload model = torch.hub.load('ultralytics/yolov5', 'yolov5s', force_reload=True)
• Notebooks – View updated notebooks
• Docker – sudo docker pull ultralytics/yolov5:latest to update your image

Thank you for spotting this issue and informing us of the problem. Please let us know if this update resolves the issue for you, and feel free to inform us of any other issues you discover or feature requests that come to mind. Happy trainings with YOLOv5 🚀!

[Bampson]

@glenn-jocher Amazing work thanks for that. This segmentation you have implemented is amazing stuff.

@WongVi Yeah a custom one, my dataset looks something like this when its ready:

dataset
├── images
│   ├── train
│   │   └── IMG_123.jpg
│   └── val
│       └── IMG_456.jpg
└── labels
    ├── train
    │   └── IMG_123.txt
    └── val
        └── IMG_456.txt

You can use https://github.com/ultralytics/JSON2YOLO to convert datasets from a bunch of different JSON formats such as COCO.
Just created a new yaml based on the COCO128-seg.yaml to fit my classes and dataset path.

[glenn-jocher]

@Bampson thank you for your kind words, but the credit truly goes to the incredible work of the YOLO community and the diligent efforts of the Ultralytics team in advancing the YOLOv5 project. We greatly appreciate your contributions and feedback, and we're delighted to hear that the segmentation functionality has been beneficial to you. Should you encounter any further questions or require assistance, please don't hesitate to reach out. Happy coding and best of luck with your custom dataset!