papers AI Learner
The Github is limit! Click to go to the new site.

SlowFast Networks for Video Recognition

2019-04-18
Christoph Feichtenhofer, Haoqi Fan, Jitendra Malik, Kaiming He

Abstract

We present SlowFast networks for video recognition. Our model involves (i) a Slow pathway, operating at low frame rate, to capture spatial semantics, and (ii) a Fast pathway, operating at high frame rate, to capture motion at fine temporal resolution. The Fast pathway can be made very lightweight by reducing its channel capacity, yet can learn useful temporal information for video recognition. Our models achieve strong performance for both action classification and detection in video, and large improvements are pin-pointed as contributions by our SlowFast concept. We report state-of-the-art accuracy on major video recognition benchmarks, Kinetics, Charades and AVA. Code will be made publicly available in PyTorch.

Abstract (translated by Google)
URL

http://arxiv.org/abs/1812.03982

PDF

http://arxiv.org/pdf/1812.03982


Similar Posts

Comments