Learning Spatio-Temporal Representation with Pseudo-3D Residual Networks
Open paper
Benchmark
Model
Rank
Results
action-recognition-in-videos-on-activitynet
P3D
#6
mAP: 78.9
action-recognition-in-videos-on-ucf101
P3D (ImageNet + Sports1M)
#47
3-fold Accuracy: 88.6
Rank counts only results with a code link.