Spatiotemporal Self-attention Modeling with Temporal Patch Shift for Action Recognition

Benchmark Model Rank Results
action-recognition-on-diving-48PSB#7Accuracy: 86
action-recognition-in-videos-on-something-something-v1TPS#10Top 1 Accuracy: 58.3
action-recognition-in-videos-on-something-something-v2TPS#39Top-1 Accuracy: 69.8
action-classification-on-kinetics-400TPS#72Acc@1: 82.5