A Implementation of SpecAugment with Tensorflow & Pytorch, introduced by Google Brain

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.

Inflated i3d network with inception backbone, weights transfered from tensorflow

Generic PyTorch dataset implementation to load and augment VIDEOS for deep learning training loops.

I3D Nonlocal ResNets in Pytorch

Code for I3D Feature Extraction

Real-world Anomaly Detection in Surveillance Videos- pytorch Re-implementation

Efficient 3D Backbone Network for Temporal Modeling

I3D features extractor with resnet50 backbone

Extract frames from videos in Python using OpenCV.

[ECCV 2024] ZeroI2V: Zero-Cost Adaptation of Pre-trained Transformers from Image to Video

An instruction to 1) download the Kinetics-400/Kinetics-600, 2) resize the videos, and 3) prepare annotations.

save a video / batch of videos as images