Video Action Transformer Network
Open paper
Benchmark
Model
Rank
Results
action-recognition-in-videos-on-ava-v21
I3D I3D
–
mAP (Val): 23.4
GFlops: 6.5
Params (M): 16.2
action-recognition-in-videos-on-ava-v21
I3D Tx HighRes
–
mAP (Val): 27.6
GFlops: 39.6
Params (M): 19.3
Rank counts only results with a code link.