https://github.com/svip-lab/svip-sequence-verification-for-procedures-in-videos
[CVPR2022] SVIP: Sequence VerIfication for Procedures in Videos
https://github.com/svip-lab/svip-sequence-verification-for-procedures-in-videos
Last synced: 17 days ago
JSON representation
[CVPR2022] SVIP: Sequence VerIfication for Procedures in Videos
- Host: GitHub
- URL: https://github.com/svip-lab/svip-sequence-verification-for-procedures-in-videos
- Owner: svip-lab
- License: mit
- Created: 2022-03-04T09:24:28.000Z (about 3 years ago)
- Default Branch: main
- Last Pushed: 2023-02-24T07:01:14.000Z (over 2 years ago)
- Last Synced: 2024-08-29T14:52:41.274Z (9 months ago)
- Language: Python
- Size: 2.26 MB
- Stars: 19
- Watchers: 2
- Forks: 2
- Open Issues: 4
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# SVIP: Sequence VerIfication for Procedures in Videos
This repo is the official implementation of our CVPR 2022 paper: [*SVIP: Sequence VerIfication for Procedures in Videos*](https://arxiv.org/abs/2112.06447).
---
### Getting Started
#### Prerequisites
- python 3.6
- pytorch 1.7.1
- cuda 10.2
#### Installation
1. Clone the repo and install dependencies.
```bash
git clone https://github.com/svip-lab/SVIP-Sequence-VerIfication-for-Procedures-in-Videos.git
cd VIP-Sequence-VerIfication-for-Procedures-in-Videos
pip install requirements.txt
```
2. Download the Kinetics-400 pretrained model.
Link：[here](https://pan.baidu.com/s/1sJU_u1QWLpeNVjymoqGO3g?pwd=bs6b)
Extraction code：bs6b
---
### Datasets
Please refer to [here](https://github.com/svip-lab/SVIP-Sequence-VerIfication-for-Procedures-in-Videos/tree/main/Datasets) for detailed instructions.
---
### Training and Evaluation
We have provided the default configuration files for reproducing our results. Try these commands to play with this project.
- For training:
```bash
CUDA_VISIBLE_DEVICES=0,1,2,3 python train.py --config configs/train_resnet_config.yml
```
- For evaluation:
```bash
CUDA_VISIBLE_DEVICES=0 python eval.py --config configs/eval_resnet_config.yml --root_path [model&log folder] --dist [L2/NormL2] --log_name [xxx]
```
Note that we use **L2** distance while evaluating on COIN-SV, otherwise **NormL2**.
---
### Trained Models
We provide checkpoints for each dataset trained with this *re-organized* codebase.
`Notice`: The reproduced performances are occassionally higher or lower (within a reasonable range) than the results reported in the paper.
DatasetSplitPaparReproduceckpt
COIN-SV
val
56.81, 0.400558.27, 0.4667here
test51.13, 0.409851.55, 0.4658
Diving48-SV
val
91.91, 1.064291.69, 1.0928here
test83.11, 0.600984.28, 0.6193
CSV
test
83.02, 0.419382.88, 0.4474here
---
### Citation
If you find this repo helpful, please cite our paper:
```
@inproceedings{qian2022svip,
title={SVIP: Sequence VerIfication for Procedures in Videos},
author={Qian, Yicheng and Luo, Weixin and Lian, Dongze and Tang, Xu and Zhao, Peilin and Gao, Shenghua},
booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
pages={19890--19902},
year={2022}
}
```