OpenGVLab / OpenGVLab/InternVideo
Evaluation on AVA-Kinetics
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 2.4k
- Forks
- 160
- PR merge metrics
- No merged PRs in 30d
Description
Hi,
I was wondering if the code for evaluating on AVA-Kinetics is available somewhere in the current codebase.
In the file 'https://github.com/OpenGVLab/InternVideo/blob/main/Downstream/Spatial-Temporal-Action-Localization/run_class_finetuning.py' L239-242, it seems that only evaluation on AVA or Kinetics separately are available.
Moreover, in the configs 'akfinetune.sh' and 'akeval.sh', --data_set is set to 'ava' and not 'ava-kinetics'.
Finally, the link to the onedrive (https://1drv.ms/u/s!AjZXNAXrK-eti2zl_JNeQbmizFbC?e=VjdMJ3) with the ava-kinetics annotations provided in the README is not working anymore.
Thanks.
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 in Downstream/Spatial-Temporal-Action-Localization/run_class_finetuning.py at lines 239-242, then compare the --data_set values in akfinetune.sh and akeval.sh. Check how AVA-Kinetics annotations are obtained from the README-linked resource. Done means the repository documents or provides a working AVA-Kinetics evaluation path and annotation source.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Feature
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 34/100