用图卷积做action recogntion
Contribution
- 提出了shift-GCN(spatial shift graph operations & temporal shift graph operations)that exceeds the state-of-the-art methods with more than 10× less computational cost,主要解决了过去工作的两个问题
- heavy computational complexity of GCN-based methods
- inflexible and predefined receptive fields of both spatial graph and temporal graph
Methods
作者灵感来源是shift-CNNs,分别就 spatial GCN 和 temporal GCN 提出新的想法,各自提了2个:
- two kinds of spatial shift graph operation
-
local shift graph operation
让physically-connected nodes 之间进行信息shift,然后再用1x1conv 来进行channel与chan
-