Details
Paper ID 50
Difficulty - Hard

Categories

  • Video recognition
  • ResNet
  • 3D Conv
  • hard

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.

Paper - https://arxiv.org/pdf/1812.03982.pdf

Dataset - https://deepmind.com/research/open-source/kinetics

https://research.google.com/ava/download.html

https://prior.allenai.org/projects/charades