Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Leveraging Spatio-Temporal Dependency for Skeleton-Based Action Recognition

About

Skeleton-based action recognition has attracted considerable attention due to its compact representation of the human body's skeletal sructure. Many recent methods have achieved remarkable performance using graph convolutional networks (GCNs) and convolutional neural networks (CNNs), which extract spatial and temporal features, respectively. Although spatial and temporal dependencies in the human skeleton have been explored separately, spatio-temporal dependency is rarely considered. In this paper, we propose the Spatio-Temporal Curve Network (STC-Net) to effectively leverage the spatio-temporal dependency of the human skeleton. Our proposed network consists of two novel elements: 1) The Spatio-Temporal Curve (STC) module; and 2) Dilated Kernels for Graph Convolution (DK-GC). The STC module dynamically adjusts the receptive field by identifying meaningful node connections between every adjacent frame and generating spatio-temporal curves based on the identified node connections, providing an adaptive spatio-temporal coverage. In addition, we propose DK-GC to consider long-range dependencies, which results in a large receptive field without any additional parameters by applying an extended kernel to the given adjacency matrices of the graph. Our STC-Net combines these two modules and achieves state-of-the-art performance on four skeleton-based action recognition benchmarks.

Jungho Lee, Minhyeok Lee, Suhwan Cho, Sungmin Woo, Sungjun Jang, Sangyoun Lee• 2022

Related benchmarks

TaskDatasetResultRank
Action RecognitionNTU RGB+D 120 (X-set)
Accuracy91.3
661
Action RecognitionNTU RGB+D (Cross-View)
Accuracy96.2
609
Action RecognitionNTU RGB+D 60 (Cross-View)
Accuracy97.1
575
Action RecognitionNTU RGB+D 60 (X-sub)--
467
Action RecognitionNTU RGB+D X-sub 120
Accuracy86.2
377
Action RecognitionNTU-120 (cross-subject (xsub))
Accuracy89.9
82
Skeleton-based Action RecognitionNW-UCLA--
44
Action RecognitionNTU-RGB+D (X-Sub)
Top-1 Acc91
23
Skeleton-based Action RecognitionNTU RGB+D X-Sub60--
16
Skeleton-based Action RecognitionNTU RGB+D X-View60
Top-1 Accuracy (E1)96.7
15
Showing 10 of 12 rows

Other info

Follow for update