An Image is Worth 16x16 Words, What is a Video Worth?

Benchmark Model Rank Results
action-recognition-in-videos-on-ucf101STAM-32 (ImageNet/Kinetics pretraining)#163-fold Accuracy: 97
action-classification-on-kinetics-400STAM (64 Frames)#92Acc@1: 80.5FLOPs (G) x views: 1040x1
action-classification-on-kinetics-400STAM (16 Frames)#106Acc@1: 79.3FLOPs (G) x views: 270x1