Pytorch for ECO Efficient Convolutional Network for Online Video Understanding [CVPR-2018]
Pytorch-version for CVPR-2015 Best paper : Fully Convolutional Networks for Semantic Segmentation
This is a general tool for video classification , detection task, this project can be used as a powerful tool to split benchmark-dataset(eg: UCF-101, THUMOS14, ActivityNet), real-scene video sets into train/val/test, maybe there are optical-flow.
paper : UntrimmedNets for Weakly Supervised Action Recognition and Detection(CVPR-2017)