Extract video features from raw videos using multiple GPUs. We support RAFT flow frames as well as S3D, I3D, R(2+1)D, VGGish, CLIP, and TIMM models.

OpenMMLab's Next Generation Video Understanding Toolbox and Benchmark

Rewriting the I3D blender addon from scratch and adding long-sought community features

Inflated i3d network with inception backbone, weights transfered from tensorflow

I3D implemetation in Keras + video preprocessing + visualization of results

PyTorch implementation of Multi-modal Dense Video Captioning (CVPR 2020 Workshops)

real-time violence detection on video content

Video Platform for Action Recognition and Object Detection in Pytorch

I3D features extractor with resnet50 backbone

Transforms for video datasets in pytorch

[CVPR 2018] Non-local Neural Networks

I3D Models in PyTorch

动作识别(Action Recognition)常见模型的Pytorch实现

A Pytorch implementation of The Visual Centrifuge: Model-Free Layered Video Representations.

Video Classification based on PyTorch

[CVPR2020] Clean-Label Backdoor Attacks on Video Recognition Models

Person-by-Person Action Recognition