3D ResNets for Action Recognition

This is the PyTorch code for the following papers:

Kensho Hara, Hirokatsu Kataoka, and Yutaka Satoh,
"Can Spatiotemporal 3D CNNs Retrace the History of 2D CNNs and ImageNet?",
arXiv preprint, arXiv:1711.09577, 2017.

Kensho Hara, Hirokatsu Kataoka, and Yutaka Satoh,
"Learning Spatio-Temporal Features with 3D Residual Networks for Action Recognition",
Proceedings of the ICCV Workshop on Action, Gesture, and Emotion Recognition, 2017.

This code includes only training and testing on the ActivityNet and Kinetics datasets.
If you want to classify your videos using our pretrained models, use this code.

The PyTorch (python) version of this code is available here.
The PyTorch version includes additional models, such as pre-activation ResNet, Wide ResNet, ResNeXt, and DenseNet.

Citation

If you use this code or pre-trained models, please cite the following:

@article{hara3dcnns,
  author={Kensho Hara and Hirokatsu Kataoka and Yutaka Satoh},
  title={Can Spatiotemporal 3D CNNs Retrace the History of 2D CNNs and ImageNet?},
  journal={arXiv preprint},
  volume={arXiv:1711.09577},
  year={2017},
}

Pre-trained models

Pre-trained models are available at releases.

Requirements

Torch

git clone https://github.com/torch/distro.git ~/torch --recursive
cd ~/torch; bash install-deps;
./install.sh

json package

luarocks install json

FFmpeg, FFprobe

wget http://johnvansickle.com/ffmpeg/releases/ffmpeg-release-64bit-static.tar.xz
tar xvf ffmpeg-release-64bit-static.tar.xz
cd ./ffmpeg-3.3.3-64bit-static/; sudo cp ffmpeg ffprobe /usr/local/bin;

Python 3

Preparation

ActivityNet

Download datasets using official crawler codes
Convert from avi to jpg files using utils/video_jpg.py

python utils/video_jpg.py avi_video_directory jpg_video_directory

Generate fps files using utils/fps.py

python utils/fps.py avi_video_directory jpg_video_directory

Kinetics

Download datasets using official crawler codes
- Locate test set in video_directory/test.
Convert from avi to jpg files using utils/video_jpg_kinetics.py

python utils/video_jpg_kinetics.py avi_video_directory jpg_video_directory

Generate n_frames files using utils/n_frames_kinetics.py

python utils/n_frames_kinetics.py jpg_video_directory

Generate annotation file in json format similar to ActivityNet using utils/kinetics_json.py

python utils/kinetics_json.py train_csv_path val_csv_path test_csv_path json_path

Running the code

Assume the structure of data directories is the following:

~/
  data/
    activitynet_videos/
      jpg/
        .../ (directories of video names)
          ... (jpg files)
    kinetics_videos/
      jpg/
        .../ (directories of class names)
          .../ (directories of video names)
            ... (jpg files)
    models/
      resnet.t7
    results/
      model_100.t7
    LR/
      ActivityNet/
        lr.lua
      Kinetics/
        lr.lua
    kinetics.json
    activitynet.json

Confirm all options.

th main.lua -h

Train ResNets-34 on the Kinetics dataset (400 classes) with 4 CPU threads (for data loading) and 2 GPUs.
Batch size is 128.
Save models at every 5 epochs.

th main.lua --root_path ~/data --video_path kinetics_videos/jpg --annotation_path kinetics.json \
--result_path results --lr_path LR/Kinetics/lr.lua --dataset kinetics --model resnet \
--resnet_depth 34 --n_classes 400 --batch_size 128 --n_gpu 2 --n_threads 4 --checkpoint 5

Continue Training from epoch 101. (~/data/results/model_100.t7 is loaded.)

th main.lua --root_path ~/data --video_path kinetics_videos/jpg --annotation_path kinetics.json \
--result_path results --lr_path LR/Kinetics/lr.lua --dataset kinetics --begin_epoch 101 \
--batch_size 128 --n_gpu 2 --n_threads 4 --checkpoint 5

Perform recognition for each video of validation set using pretrained model. This operation outputs top-10 labels for each video.

th main.lua --root_path ~/data --video_path kinetics_videos/jpg --annotation_path kinetics.json \
--result_path results --premodel_path models/resnet.t7 --dataset kinetics \
--no_train --no_val --test_video --test_subset val --n_gpu 2 --n_threads 4

Chandanpanda/3D-ResNets