End-to-End Learning of Motion Representation for Video Understanding
Open paper
Benchmark
Model
Rank
Results
action-recognition-in-videos-on-ucf101
TVNet+IDT
#25
3-fold Accuracy: 95.4
action-recognition-in-videos-on-hmdb-51
TVNet+IDT
#31
Average accuracy of 3 splits: 72.6
Rank counts only results with a code link.