Richly Activated Graph Convolutional Network for Action Recognition with Incomplete Skeletons

16 May 2019Yi-Fan SongZhang ZhangLiang Wang

Current methods for skeleton-based human action recognition usually work with completely observed skeletons. However, in real scenarios, it is prone to capture incomplete and noisy skeletons, which will deteriorate the performance of traditional models... (read more)

PDF Abstract
TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT BENCHMARK
Skeleton Based Action Recognition NTU RGB+D 2s RA-GCN Accuracy (CV) 93.0 # 23
Accuracy (CS) 85.8 # 28
Skeleton Based Action Recognition NTU RGB+D 3s RA-GCN Accuracy (CV) 93.5 # 19
Accuracy (CS) 85.9 # 27

Methods used in the Paper