Yolov5: How to get the related class labels from Mosaic augmentation?

[innat]

I'm trying to get the related class label from mosaic augmentation for multi-class object detection. I'm not using Yolo5 but the mosaic implementation in my case inspired by Yolo implementation. I described my query in more detail over here, SO. Please, have a look.

[github-actions[bot]]

Hello @innat, thank you for your interest in our work! Please visit our Custom Training Tutorial to get started, and see our Jupyter Notebook , Docker Image, and Google Cloud Quickstart Guide for example environments.

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 model or data training question, please note Ultralytics does not provide free personal support. As a leader in vision ML and AI, we do offer professional consulting, from simple expert advice up to delivery of fully customized, end-to-end production solutions for our clients, such as:

• Cloud-based AI systems operating on hundreds of HD video streams in realtime.
• Edge AI integrated into custom iOS and Android apps for realtime 30 FPS video inference.
• Custom data training, hyperparameter evolution, and model exportation to any destination.

For more information please visit https://www.ultralytics.com.

[glenn-jocher]

@innat if you are looking for multi-class object detection, I would simply use YOLOv5, which already incorporates all the mosaic augmentation you want.

[innat]

@glenn-jocher Could you please point me to the specific code snippet where the mosaic augmentation in Yolo5 deals with such cases? So, that I can understand when it randomly making a mosaic, how it tracks the labels in final results.

[glenn-jocher]

@innat of course. load_mosaic() in datasets.py is responsible for all mosaic activity. Your dataset trains with Mosaic by default in YOLOv5, no extra action is needed on your part, simply python train.py --data custom.yaml

https://github.com/ultralytics/yolov5/blob/4d3680c81dad760e5d1ffe397ab2fc9169d9fb70/utils/datasets.py#L643-L699

[glenn-jocher]

You can also view your mosaics during training. Suggest you start with the custom training tutorial, it explains all of this already.
https://github.com/ultralytics/yolov5#tutorials

[innat]

Thanks for the pointer and suggestion. -)

[innat]

The problem is solved. Here is the answer, leaving it here for future readers.