Skeleton Image Representation for 3D Action Recognition based on Tree Structure and Reference Joints

11 Sep 2019  ·  Carlos Caetano, François Brémond, William Robson Schwartz ·

In the last years, the computer vision research community has studied on how to model temporal dynamics in videos to employ 3D human action recognition. To that end, two main baseline approaches have been researched: (i) Recurrent Neural Networks (RNNs) with Long-Short Term Memory (LSTM); and (ii) skeleton image representations used as input to a Convolutional Neural Network (CNN). Although RNN approaches present excellent results, such methods lack the ability to efficiently learn the spatial relations between the skeleton joints. On the other hand, the representations used to feed CNN approaches present the advantage of having the natural ability of learning structural information from 2D arrays (i.e., they learn spatial relations from the skeleton joints). To further improve such representations, we introduce the Tree Structure Reference Joints Image (TSRJI), a novel skeleton image representation to be used as input to CNNs. The proposed representation has the advantage of combining the use of reference joints and a tree structure skeleton. While the former incorporates different spatial relationships between the joints, the latter preserves important spatial relations by traversing a skeleton tree with a depth-first order algorithm. Experimental results demonstrate the effectiveness of the proposed representation for 3D action recognition on two datasets achieving state-of-the-art results on the recent NTU RGB+D~120 dataset.

PDF Abstract

Results from the Paper


Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Skeleton Based Action Recognition NTU RGB+D TSRJI (Late Fusion) + HCN Accuracy (CV) 80.3 # 109
Accuracy (CS) 73.3 # 111
Action Recognition NTU RGB+D 120 TSRJI Accuracy (Cross-Subject) 62.8 # 17
Accuracy (Cross-Setup) 67.9 # 15
Skeleton Based Action Recognition NTU RGB+D 120 TSRJI (Late Fusion) Accuracy (Cross-Subject) 65.5% # 57
Accuracy (Cross-Setup) 59.7% # 64
Skeleton Based Action Recognition NTU RGB+D 120 TSRJI (Late Fusion) + HCN Accuracy (Cross-Subject) 67.9% # 55
Accuracy (Cross-Setup) 62.8% # 60

Methods


No methods listed for this paper. Add relevant methods here