facebookresearch/co-tracker
CoTracker is a model for tracking any point (pixel) on a video.
Jupyter NotebookNOASSERTION
Issues
- 0
The huggingface demo doesn't work.
#90 opened - 3
- 1
- 1
- 2
About reproducing the paper
#86 opened - 1
- 0
huggingface space down?
#82 opened - 1
- 3
Sample occluded points during training
#79 opened - 5
- 2
how to track a specified x-y point?
#77 opened - 3
question about windowed inference
#74 opened - 0
thx
#73 opened - 1
track fast moving objects
#72 opened - 8
- 2
- 2
- 2
Efficency of V2 and adding more points?
#67 opened - 1
question about backward_tracking
#66 opened - 8
- 2
Tracking errors with cotracker2
#64 opened - 1
About reproduction of training.
#63 opened - 1
- 4
- 3
Inference with sequence of multiple images for a particular single pixel using co-tracker2
#59 opened - 3
Solve camera poses?
#58 opened - 4
- 8
question about v2
#54 opened - 0
Chinese newbie seeking help! Encountering issues while trying to install libraries using pip!
#52 opened - 9
- 3
- 1
Input shapes in forward_batch()
#47 opened - 5
- 1
- 5
Specular reflective surfaces
#44 opened - 2
- 4
- 2
- 7
How to prepare new dataset?
#39 opened - 2
RuntimeError: Invalid buffer size
#38 opened - 19
CoTracker application on live videos
#37 opened - 10
Training with different batch size
#36 opened - 1
missing re-trace capability
#35 opened - 3
- 2
- 6
Is a relicensing planned like Dinov2?
#31 opened - 4
moviepy saving video error
#30 opened - 3
Error when loading model using torch.hub
#29 opened - 4
Occluded points at t = `query_frame`
#28 opened - 4