The official pytorch implementation of our paper "Is Space-Time Attention All You Need for Video Understanding?"

Overview

TimeSformer

This is an official pytorch implementation of Is Space-Time Attention All You Need for Video Understanding?. In this repository, we provide PyTorch code for training and testing our proposed TimeSformer model. TimeSformer provides an efficient video classification framework that achieves state-of-the-art results on several video action recognition benchmarks such as Kinetics-400.

If you find TimeSformer useful in your research, please use the following BibTeX entry for citation.

@misc{bertasius2021spacetime,
    title   = {Is Space-Time Attention All You Need for Video Understanding?},
    author  = {Gedas Bertasius and Heng Wang and Lorenzo Torresani},
    year    = {2021},
    eprint  = {2102.05095},
    archivePrefix = {arXiv},
    primaryClass = {cs.CV}
}

Model Zoo

We provide TimeSformer models pretrained on Kinetics-400 (K400), Kinetics-600 (K600), Something-Something-V2 (SSv2), and HowTo100M datasets.

name dataset # of frames spatial crop [email protected] [email protected] url
TimeSformer K400 8 224 77.9 93.2 model
TimeSformer-HR K400 16 448 79.6 94.0 model
TimeSformer-L K400 96 224 80.6 94.7 model
name dataset # of frames spatial crop [email protected] [email protected] url
TimeSformer K600 8 224 79.1 94.4 model
TimeSformer-HR K600 16 448 81.8 95.8 model
TimeSformer-L K600 96 224 82.2 95.6 model
name dataset # of frames spatial crop [email protected] [email protected] url
TimeSformer SSv2 8 224 59.1 85.6 model
TimeSformer-HR SSv2 16 448 61.8 86.9 model
TimeSformer-L SSv2 64 224 62.0 87.5 model
name dataset # of frames spatial crop single clip coverage [email protected] url
TimeSformer HowTo100M 8 224 8.5s 56.8 model
TimeSformer HowTo100M 32 224 34.1s 61.2 model
TimeSformer HowTo100M 64 448 68.3s 62.2 model
TimeSformer HowTo100M 96 224 102.4s 62.6 model

We note that these models were retrained using a slightly different implementation than the one used in the paper. Therefore, there might be a small difference in performance compared to the results reported in the paper.

Installation

First, create a conda virtual environment and activate it:

conda create -n timesformer python=3.7 -y
source activate timesformer

Then, install the following packages:

  • torchvision: pip install torchvision or conda install torchvision -c pytorch
  • fvcore: pip install 'git+https://github.com/facebookresearch/fvcore'
  • simplejson: pip install simplejson
  • einops: pip install einops
  • timm: pip install timm
  • PyAV: conda install av -c conda-forge
  • psutil: pip install psutil
  • OpenCV: pip install opencv-python
  • tensorboard: pip install tensorboard

Lastly, build the TimeSformer codebase by running:

git clone https://github.com/facebookresearch/TimeSformer
cd TimeSformer
python setup.py build develop

Usage

Dataset Preparation

Please use the dataset preparation instructions provided in DATASET.md.

Training the Default TimeSformer

Training the default TimeSformer that uses divided space-time attention, and operates on 8-frame clips cropped at 224x224 spatial resolution, can be done using the following command:

python tools/run_net.py \
  --cfg configs/Kinetics/TimeSformer_divST_8x32_224.yaml \
  DATA.PATH_TO_DATA_DIR path_to_your_dataset \
  NUM_GPUS 8 \
  TRAIN.BATCH_SIZE 8 \

You may need to pass location of your dataset in the command line by adding DATA.PATH_TO_DATA_DIR path_to_your_dataset, or you can simply add

DATA:
  PATH_TO_DATA_DIR: path_to_your_dataset

To the yaml configs file, then you do not need to pass it to the command line every time.

Using a Different Number of GPUs

If you want to use a smaller number of GPUs, you need to modify .yaml configuration files in configs/. Specifically, you need to modify the NUM_GPUS, TRAIN.BATCH_SIZE, TEST.BATCH_SIZE, DATA_LOADER.NUM_WORKERS entries in each configuration file. The BATCH_SIZE entry should be the same or higher as the NUM_GPUS entry. In configs/Kinetics/TimeSformer_divST_8x32_224_4gpus.yaml, we provide a sample configuration file for a 4 GPU setup.

Using Different Self-Attention Schemes

If you want to experiment with different space-time self-attention schemes, e.g., space-only or joint space-time attention, use the following commands:

python tools/run_net.py \
  --cfg configs/Kinetics/TimeSformer_spaceOnly_8x32_224.yaml \
  DATA.PATH_TO_DATA_DIR path_to_your_dataset \
  NUM_GPUS 8 \
  TRAIN.BATCH_SIZE 8 \

and

python tools/run_net.py \
  --cfg configs/Kinetics/TimeSformer_jointST_8x32_224.yaml \
  DATA.PATH_TO_DATA_DIR path_to_your_dataset \
  NUM_GPUS 8 \
  TRAIN.BATCH_SIZE 8 \

Training Different TimeSformer Variants

If you want to train more powerful TimeSformer variants, e.g., TimeSformer-HR (operating on 16-frame clips sampled at 448x448 spatial resolution), and TimeSformer-L (operating on 96-frame clips sampled at 224x224 spatial resolution), use the following commands:

python tools/run_net.py \
  --cfg configs/Kinetics/TimeSformer_divST_16x16_448.yaml \
  DATA.PATH_TO_DATA_DIR path_to_your_dataset \
  NUM_GPUS 8 \
  TRAIN.BATCH_SIZE 8 \

and

python tools/run_net.py \
  --cfg configs/Kinetics/TimeSformer_divST_96x4_224.yaml \
  DATA.PATH_TO_DATA_DIR path_to_your_dataset \
  NUM_GPUS 8 \
  TRAIN.BATCH_SIZE 8 \

Note that for these models you will need a set of GPUs with ~32GB of memory.

Inference

Use TRAIN.ENABLE and TEST.ENABLE to control whether training or testing is required for a given run. When testing, you also have to provide the path to the checkpoint model via TEST.CHECKPOINT_FILE_PATH.

python tools/run_net.py \
  --cfg configs/Kinetics/TimeSformer_divST_8x32_224_TEST.yaml \
  DATA.PATH_TO_DATA_DIR path_to_your_dataset \
  TEST.CHECKPOINT_FILE_PATH path_to_your_checkpoint \
  TRAIN.ENABLE False \

Single-Node Training via Slurm

To train TimeSformer via Slurm, please check out our single node Slurm training script slurm_scripts/run_single_node_job.sh.

Multi-Node Training via Submitit

Distributed training is available via Slurm and submitit

pip install submitit

To train TimeSformer model on Kinetics using 4 nodes with 8 gpus each use the following command:

python tools/submit.py --cfg configs/Kinetics/TimeSformer_divST_8x32_224.yaml --job_dir  /your/job/dir/${JOB_NAME}/ --num_shards 4 --name ${JOB_NAME} --use_volta32

We provide a script for launching slurm jobs in slurm_scripts/run_multi_node_job.sh.

Finetuning

To finetune from an existing PyTorch checkpoint add the following line in the command line, or you can also add it in the YAML config:

TRAIN.CHECKPOINT_FILE_PATH path_to_your_PyTorch_checkpoint
TRAIN.FINETUNE True

HowTo100M Dataset Split

If you want to experiment with the long-term video modeling task on HowTo100M, please download the train/test split files from here.

Environment

The code was developed using python 3.7 on Ubuntu 20.04. For training, we used four GPU compute nodes each node containing 8 Tesla V100 GPUs (32 GPUs in total). Other platforms or GPU cards have not been fully tested.

License

The majority of this work is licensed under CC-NC 4.0 International license. However portions of the project are available under separate license terms: SlowFast and pytorch-image-models are licensed under the Apache 2.0 license.

Contributing

We actively welcome your pull requests. Please see CONTRIBUTING.md and CODE_OF_CONDUCT.md for more info.

Acknowledgements

TimeSformer is built on top of PySlowFast and pytorch-image-models by Ross Wightman. We thank the authors for releasing their code. If you use our model, please consider citing these works as well:

@misc{fan2020pyslowfast,
  author =       {Haoqi Fan and Yanghao Li and Bo Xiong and Wan-Yen Lo and
                  Christoph Feichtenhofer},
  title =        {PySlowFast},
  howpublished = {\url{https://github.com/facebookresearch/slowfast}},
  year =         {2020}
}
@misc{rw2019timm,
  author = {Ross Wightman},
  title = {PyTorch Image Models},
  year = {2019},
  publisher = {GitHub},
  journal = {GitHub repository},
  doi = {10.5281/zenodo.4414861},
  howpublished = {\url{https://github.com/rwightman/pytorch-image-models}}
}
Owner
Facebook Research
Facebook Research
Custom TensorFlow2 implementations of forward and backward computation of soft-DTW algorithm in batch mode.

Batch Soft-DTW(Dynamic Time Warping) in TensorFlow2 including forward and backward computation Custom TensorFlow2 implementations of forward and backw

19 Aug 30, 2022
Demonstrational Session git repo for H SAF User Workshop (28/1)

5th H SAF User Workshop The 5th H SAF User Workshop supported by EUMeTrain will be held in online in January 24-28 2022. This repository contains inst

H SAF 4 Aug 04, 2022
Generalized Jensen-Shannon Divergence Loss for Learning with Noisy Labels

The official code for the NeurIPS 2021 paper Generalized Jensen-Shannon Divergence Loss for Learning with Noisy Labels

13 Dec 22, 2022
Denoising Diffusion Implicit Models

Denoising Diffusion Implicit Models (DDIM) Jiaming Song, Chenlin Meng and Stefano Ermon, Stanford Implements sampling from an implicit model that is t

465 Jan 05, 2023
Deep-Learning-Image-Captioning - Implementing convolutional and recurrent neural networks in Keras to generate sentence descriptions of images

Deep Learning - Image Captioning with Convolutional and Recurrent Neural Nets ========================================================================

23 Apr 06, 2022
YOLOX is a high-performance anchor-free YOLO, exceeding yolov3~v5 with ONNX, TensorRT, ncnn, and OpenVINO supported.

Introduction YOLOX is an anchor-free version of YOLO, with a simpler design but better performance! It aims to bridge the gap between research and ind

7.7k Jan 03, 2023
Official repository of the paper Privacy-friendly Synthetic Data for the Development of Face Morphing Attack Detectors

SMDD-Synthetic-Face-Morphing-Attack-Detection-Development-dataset Official repository of the paper Privacy-friendly Synthetic Data for the Development

10 Dec 12, 2022
Benchmark for evaluating open-ended generation

OpenMEVA Contributed by Jian Guan, Zhexin Zhang. Thank Jiaxin Wen for DeBugging. OpenMEVA is a benchmark for evaluating open-ended story generation me

25 Nov 15, 2022
This project provides the code and datasets for 'CapSal: Leveraging Captioning to Boost Semantics for Salient Object Detection', CVPR 2019.

Code-and-Dataset-for-CapSal This project provides the code and datasets for 'CapSal: Leveraging Captioning to Boost Semantics for Salient Object Detec

lu zhang 48 Aug 19, 2022
The Python ensemble sampling toolkit for affine-invariant MCMC

emcee The Python ensemble sampling toolkit for affine-invariant MCMC emcee is a stable, well tested Python implementation of the affine-invariant ense

Dan Foreman-Mackey 1.3k Dec 31, 2022
[Open Source]. The improved version of AnimeGAN. Landscape photos/videos to anime

[Open Source]. The improved version of AnimeGAN. Landscape photos/videos to anime

CC 4.4k Dec 27, 2022
Python library for loading and using triangular meshes.

Trimesh is a pure Python (2.7-3.4+) library for loading and using triangular meshes with an emphasis on watertight surfaces. The goal of the library i

Michael Dawson-Haggerty 2.2k Jan 07, 2023
Aalto-cs-msc-theses - Listing of M.Sc. Theses of the Department of Computer Science at Aalto University

Aalto-CS-MSc-Theses Listing of M.Sc. Theses of the Department of Computer Scienc

Jorma Laaksonen 3 Jan 27, 2022
Pytorch implementation of MalConv

MalConv-Pytorch A Pytorch implementation of MalConv Desciprtion This is the implementation of MalConv proposed in Malware Detection by Eating a Whole

Alexander H. Liu 58 Oct 26, 2022
Cancer-and-Tumor-Detection-Using-Inception-model - In this repo i am gonna show you how i did cancer/tumor detection in lungs using deep neural networks, specifically here the Inception model by google.

Cancer-and-Tumor-Detection-Using-Inception-model In this repo i am gonna show you how i did cancer/tumor detection in lungs using deep neural networks

Deepak Nandwani 1 Jan 01, 2022
This is official implementaion of paper "Token Shift Transformer for Video Classification".

This is official implementaion of paper "Token Shift Transformer for Video Classification". We achieve SOTA performance 80.40% on Kinetics-400 val. Paper link

VideoNet 60 Dec 30, 2022
Using LSTM to detect spoofing attacks in an Air-Ground network

Using LSTM to detect spoofing attacks in an Air-Ground network Specifications IDE: Spider Packages: Tensorflow 2.1.0 Keras NumPy Scikit-learn Matplotl

Tiep M. H. 1 Nov 20, 2021
Independent and minimal implementations of some reinforcement learning algorithms using PyTorch (including PPO, A3C, A2C, ...).

PyTorch RL Minimal Implementations There are implementations of some reinforcement learning algorithms, whose characteristics are as follow: Less pack

Gemini Light 4 Dec 31, 2022
MediaPipe is a an open-source framework from Google for building multimodal

MediaPipe is a an open-source framework from Google for building multimodal (eg. video, audio, any time series data), cross platform (i.e Android, iOS, web, edge devices) applied ML pipelines. It is

Bhavishya Pandit 3 Sep 30, 2022
Hierarchical Metadata-Aware Document Categorization under Weak Supervision (WSDM'21)

Hierarchical Metadata-Aware Document Categorization under Weak Supervision This project provides a weakly supervised framework for hierarchical metada

Yu Zhang 53 Sep 17, 2022