search

cvat-ai / cvat

Annotate better with CVAT, the industry-leading data engine for machine learning. Used and trusted by teams at any scale, for data of any scale.
https://cvat.ai
MIT License
12.31k stars 2.96k forks source link

Didn't receive all labels for images when downloading dataset in YOLOv8 detection format #8450

Open benjiroooo opened 1 week ago

benjiroooo commented 1 week ago

Actions before raising this issue

Steps to Reproduce

  1. Right click on project
  2. Click export dataset
  3. Export in YOLOv8 detection format

Expected Behavior

Receive all labels for images, only received half of them.

Possible Solution

You can export as YOLOv8 Oriented Bounding Boxes, giving you all the labels, but the label format is different. Also, the labels don't align with the annotated image it's attached to. e.g. frame245 has 2 shapes while the label only has one shape.

Context

I can't train a model with incomplete data, I have 211 images and only received 115 labels.

Environment

OS: Windows 11
bsekachev commented 1 week ago

Please, provide a backup with a task where the problem may be minimally reproduced.