kinetics400

Awesome video understanding toolkits based on PaddlePaddle. It supports video data annotation tools, lightweight RGB and skeleton based action recognition model, practical applications for video tagging and sport action detection.

[AAAI 2023 (Oral)] CrissCross: Self-Supervised Audio-Visual Representation Learning with Relaxed Cross-Modal Synchronicity

[AAAI 2024] XKD: Cross-modal Knowledge Distillation with Domain Alignment for Video Representation Learning.

A Human Action Recognition pipeline using MMAction2 and kinetics400 dataset. MMAction2 is an open-source toolbox for video understanding based on PyTorch.

IE Fitizens Corporate Project