Contrastive Learning from Extremely Augmented Skeleton Sequences for Self-supervised Action Recognition
Open paper
Benchmark
Model
Rank
Results
self-supervised-human-action-recognition-on-ntu-rgbd-120
3s-AimCLR
#2
xsub (%): 68.2
xset (%): 68.8
Encoder: ST-GCN
Classifier: FC
Rank counts only results with a code link.