facebookresearch / facebookresearch/co-tracker
Inaccurate tracking of animal's legs
Open
- Dominant language
- Jupyter Notebook
- Stars
- 5.1k
- Forks
- 389
- PR merge metrics
- No merged PRs in 30d
Description
Hi, I tried this work for some videos of moving animals (https://www.youtube.com/watch?v=DaYXX-68jSM). I found the model's prediction fits the body well but works poorly for legs. The tracking points of two legs often changes to those of one leg. Is there a good solution here? Thanks you!
Contributor guide
Research direction
No files, tests, or entry points are identified. Start by reproducing the behavior with the linked animal video and inspect how the model tracks the legs versus the body. Done means the two leg points remain distinct and follow the correct legs throughout the video.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- jupyter-notebook
- Domain
- computer-vision, machine-learning
- Issue type
- Bug
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100