RSPNet
Official Pytorch implementation for AAAI2021 paper "RSPNet: Relative Speed Perception for Unsupervised Video Representation Learning"
Getting Started
Install Dependencies
All dependencies can be installed using pip:
python -m pip install -r requirements.txt
Our experiments run on Python 3.7 and PyTorch 1.6. Other versions should work but are not tested.
Transcode Videos (Optional)
This step is optional but will increase the data loading speed dramatically.
We decode the videos on the fly while training so we don't need to split frames. This makes disk IO a lot faster but increases CPU usage. This transcode step aims at reducing CPU consumed by decoding by 1) lower video resolution. 2) add more key frames.
To perform transcode, you need to have ffmpeg installed, then run:
python utils/transcode_dataset.py PATH/TO/ORIGIN_VIDEOS PATH/TO/TRANSCODED_VIDEOS
Be warned, this will use all your CPU and will take several hours (on our Intel E5-2630 *2 workstation) to complete.
Prepare Datasets
Your are expected to prepare date for pre-training (Kinetics-400 dataset) and fine-tuning (UCF101, HMDB51 and Something-something-v2 datasets). To let the scripts find datasets on your system, the recommended way is to create symbolic links in ./data directory to the actual path. We found this solution flexible.
The expected directory hierarchy is as follow:
├── data
│ ├── hmdb51
│ │ ├── metafile
│ │ │ ├── brush_hair_test_split1.txt
│ │ │ └── ...
│ │ └── videos
│ │ ├── brush_hair
│ │ │ └── *.avi
│ │ └── ...
│ ├── UCF101
│ │ ├── ucfTrainTestlist
│ │ │ ├── classInd.txt
│ │ │ ├── testlist01.txt
│ │ │ ├── trainlist01.txt
│ │ │ └── ...
│ │ └── UCF-101
│ │ ├── ApplyEyeMakeup
│ │ │ └── *.avi
│ │ └── ...
│ ├── kinetics400
│ │ ├── train_video
│ │ │ ├── answering_questions
│ │ │ │ └── *.mp4
│ │ │ └── ...
│ │ └── val_video
│ │ └── (same as train_video)
│ ├── kinetics100
│ │ └── (same as kinetics400)
│ └── smth-smth-v2
│ ├── 20bn-something-something-v2
│ │ └── *.mp4
│ └── annotations
│ ├── something-something-v2-labels.json
│ ├── something-something-v2-test.json
│ ├── something-something-v2-train.json
│ └── something-something-v2-validation.json
└── ...
Alternatively, you can change the path in config/dataset to match your system.
Build Kinetics-100 dataset (Optional)
Some of our ablation study experiments use the Kinetics-100 dataset for pre-training. This dataset is built by extract 100 classes from Kinetics-400, which has the smallest file size on the train set.
If you have Kinetics-400 available, you can build Kinetics-100 by:
python -m utils.build_kinetics_subset
This script will create symbolic links instead of copy data. It is expected to complete in a minute.
We have included a pre-built one at data/kinetics100_links and created the symbolic link data/kinetics100 that related to it. You need to have data/kinetics400 available at runtime.
Pre-training on Pretext Tasks
Now you have set up the environment. Run the following command to pre-train your models on pretext tasks.
export CUDA_VISIBLE_DEVICES=0,1,2,3
# Architecture: C3D
python pretrain.py -e exps/pretext-c3d -c config/pretrain/c3d.jsonnet
# Architecture: ResNet-18
python pretrain.py -e exps/pretext-resnet18 -c config/pretrain/resnet18.jsonnet
# Architecture: S3D-G
python pretrain.py -e exps/pretext-s3dg -c config/pretrain/s3dg.jsonnet
# Architecture: R(2+1)D
python pretrain.py -e exps/pretext-r2plus1d -c config/pretrain/r2plus1d.jsonnet
You can use kinetics100 dataset for training by editing config/pretrain/moco-train-base.jsonnet (line 13)
Action Recognition
After pre-trained on pretext tasks, these models are fine-tuned to perform action recognition task on UCF101, HMDB51 and Something-something-v2 datasets.
export CUDA_VISIBLE_DEVICES=0,1
# Dataset: UCF101
# Architecture: C3D [email protected]=76.71%
python finetune.py -c config/finetune/ucf101_c3d.jsonnet \
--mc exps/pretext-c3d/model_best.pth.tar \
-e exps/ucf101-c3d
# Architecture: ResNet-18 [email protected]=74.33%
python finetune.py -c config/finetune/ucf101_resnet18.jsonnet \
--mc exps/pretext-resnet18/model_best.pth.tar \
-e exps/ucf101-resnet18
# Architecture: S3D-G [email protected]=89.9%
python finetune.py -c config/finetune/ucf101_s3dg.jsonnet \
--mc exps/pretext-s3dg/model_best.pth.tar \
-e exps/ucf101-s3dg
# Architecture: R(2+1)D [email protected]=81.1%
python finetune.py -c config/finetune/ucf101_r2plus1d.jsonnet \
--mc exps/pretext-r2plus1d/model_best.pth.tar \
-e exps/ucf101-r2plus1d
# Dataset: HMDB51
# Architecture: C3D [email protected]=44.58%
python finetune.py -c config/finetune/hmdb51_c3d.jsonnet \
--mc exps/pretext-c3d/model_best.pth.tar \
-e exps/hmdb51-c3d
# Architecture: ResNet-18 [email protected]=41.83%
python finetune.py -c config/finetune/hmdb51_resnet18.jsonnet \
--mc exps/pretext-resnet18/model_best.pth.tar \
-e exps/hmdb51-resnet18
# Architecture: S3D-G [email protected]=59.6%
python finetune.py -c config/finetune/hmdb51_s3dg.jsonnet \
--mc exps/pretext-s3dg/model_best.pth.tar \
-e exps/hmdb51-s3dg
# Architecture: R(2+1)D [email protected]=44.6%
python finetune.py -c config/finetune/hmdb51_r2plus1d.jsonnet \
--mc exps/pretext-r2plus1d/model_best.pth.tar \
-e exps/hmdb51-r2plus1d
# Dataset: Something-something-v2
# Architecture: C3D [email protected]=47.76%
python finetune.py -c config/finetune/smth_smth_c3d.jsonnet \
--mc exps/pretext-c3d/model_best.pth.tar \
-e exps/smthv2-c3d
# Architecture: ResNet-18 [email protected]=44.02%
python finetune.py -c config/finetune/smth_smth_resnet18.jsonnet \
--mc exps/pretext-resnet18/model_best.pth.tar \
-e exps/smthv2-resnet18
# Architecture: S3D-G [email protected]=55.03%
python finetune.py -c config/finetune/smth_smth_s3dg.jsonnet \
--mc exps/pretext-s3dg/model_best.pth.tar \
-e exps/smthv2-s3dg
Results and Pre-trained Models
| Architecture | Pre-trained dataset | Pre-training epoch | Pre-trained model | Acc. on UCF101 | Acc. on HMDB51 |
|---|---|---|---|---|---|
| S3D-G | Kinetics-400 | 1000 | Download link | 93.7 | 64.7 |
| S3D-G | Kinetics-400 | 200 | Download link | 89.9 | 59.6 |
| R(2+1)D | Kinetics-400 | 200 | Download link | 81.1 | 44.6 |
| ResNet-18 | Kinetics-400 | 200 | Download link | 74.3 | 41.8 |
| C3D | Kinetics-400 | 200 | Download link | 76.7 | 44.6 |
Video Retrieval
The pretrained model can also be used in searching relevant videos based on the given query video.
export CUDA_VISIBLE_DEVICES=0 # use single GPU
python retrieval.py -c config/retrieval/ucf101_resnet18.jsonnet \
--mc exps/pretext-resnet18/model_best.pth.tar \
-e exps/retrieval-resnet18
The video retrieval result in our paper
| Architecture | k=1 | k=5 | k=10 | k=20 | k=50 |
|---|---|---|---|---|---|
| C3D | 36.0 | 56.7 | 66.5 | 76.3 | 87.7 |
| ResNet-18 | 41.1 | 59.4 | 68.4 | 77.8 | 88.7 |
Visualization
We further visualize the region of interest (RoI) that contributes most to the similarity score using the class activation map (CAM) technique.
export CUDA_VISIBLE_DEVICES=0,1
python visualization.py -c config/pretrain/s3dg.jsonnet \
--load-model exps/pretext-s3dg/model_best.pth.tar \
-e exps/visual-s3dg \
-x '{batch_size: 1}'
The cam visualization results will be plotted in png files like
Troubleshoot
-
DECORDError cannot find video stream with wanted index: -1
Some video from Kinetics dataset does not contain a valid video stream for some unknown reason. To filter them out, run
python utils/verify_video.py PATH/TO/VIDEOS, then copy the output to theblacklistconfig inconfig/dataset/kinetics{400,100}.libsonnet. You need to haveffmpeginstalled.
Citation
Please cite the following paper if you feel RSPNet useful to your research
@InProceedings{chen2020RSPNet,
author = {Peihao Chen, Deng Huang, Dongliang He, Xiang Long, Runhao Zeng, Shilei Wen, Mingkui Tan, and Chuang Gan},
title = {RSPNet: Relative Speed Perception for Unsupervised Video Representation Learning},
booktitle = {The AAAI Conference on Artificial Intelligence (AAAI)},
year = {2021}
}
Contact
For any question, please file an issue or contact
Peihao Chen: [email protected]
Deng Huang: [email protected]
30 Sep 07, 2021
236 Dec 21, 2022
95 Jan 08, 2023
421 Jan 04, 2023
14 Dec 25, 2022
33 Oct 12, 2022
276 Dec 20, 2022
64 Jan 07, 2023
47 Jan 09, 2023
585 Oct 12, 2022
1 May 08, 2022
190 Dec 27, 2022
29 Jan 06, 2023
44 Dec 06, 2022
14 Jul 24, 2022
14 Jan 04, 2023
15 Jul 11, 2022
869 Jan 07, 2023
34 Jul 22, 2022
1 Jan 30, 2022