Tile (slice) YOLO Dataset for Small Objects Detection

[jehan88]

hello all, i have question about Tile (slice) YOLO Dataset for Small Objects Detection, i should do annotate for the image then i tile the image and txt? my secund question :can i use the tiling result with yolov5 to train the model with the tiling image result?

[github-actions[bot]]

Hello @jehan88, thank you for your interest in our work! Ultralytics has publicly released YOLOv5 at https://github.com/ultralytics/yolov5, featuring faster, lighter and more accurate object detection. YOLOv5 is recommended for all new projects.

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 Glenn Jocher at glenn.jocher@ultralytics.com.

[github-actions[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[glenn-jocher]

@jehan88 hello! To tile (slice) a YOLO dataset for small object detection, you should first annotate the original images and then tile both the images and their corresponding annotation files. Once you have the tiled dataset, you can use it to train a YOLOv5 model without any issues. Just ensure that your annotations are adjusted to the tiled images accordingly.

[glenn-jocher]

@jehan88 yes, you can tile (slice) the YOLO dataset for small objects detection by annotating the original image and then dividing it into smaller tiles. You can also create corresponding annotation files for each tile.

You can use the tiling result with YOLOv5 to train the model by modifying the data loading code to accommodate the tiled images and annotations. You can refer to the YOLOv5 documentation for guidance on how to implement this.