Action Recognition with Trajectory-Pooled Deep-Convolutional Descriptors
Open paper
Benchmark
Model
Rank
Results
action-recognition-in-videos-on-hmdb-51
TDD + IDT
#40
Average accuracy of 3 splits: 65.9
action-recognition-in-videos-on-ucf101
TDD + IDT
#43
3-fold Accuracy: 91.5
Rank counts only results with a code link.