Two-Stream Video Classification with Cross-Modality Attention
Open paper
Benchmark
Model
Rank
Results
action-classification-on-kinetics-400
CMA iter1 (16 frames)
–
Acc@1: 75.98
action-recognition-in-videos-on-ucf101
CMA iter1-S
–
3-fold Accuracy: 96.5
Rank counts only results with a code link.