OpenGVLab / OpenGVLab/InternVideo
NEED HELP: Action Classification low performance
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 2.4k
- Forks
- 160
- PR merge metrics
- No merged PRs in 30d
Description
Problems: Use demo to test action classification on kinetics-700 validation set but get very poor result
Experiment:
- Pretrained model: https://huggingface.co/OpenGVLab/InternVideo2-Stage2_1B-224p-f4/tree/main
- text candidate：use the class name of k700 dataset annotation
- dataset：kinetics-700 validation set.
- code：demo.ipynb
example：
input：carving ice/nTnAoTQ41Nc_000011_000021.mp4
output：
text: coloring in ~ prob: 0.0085
text: acting in play ~ prob: 0.0053
text: smashing ~ prob: 0.0043
text: cracking knuckles ~ prob: 0.0041
text: tasting food ~ prob: 0.0040
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by running demo.ipynb with the listed InternVideo2-Stage2_1B-224p-f4 checkpoint, Kinetics-700 validation data, and class-name text candidates. Reproduce the carving-ice example, then trace the demo inputs and outputs to identify why the predicted probabilities do not match the expected action; done means the discrepancy has a reproducible explanation or correction.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- huggingface, jupyter-notebook, python
- Domain
- computer-vision, machine-learning
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100