View Adaptive Recurrent Neural Networks for High Performance Human Action Recognition from Skeleton Data

TL;DR

This paper introduces a view adaptive RNN that automatically adjusts observation viewpoints during human action recognition from skeleton data, significantly improving consistency and accuracy across different views.

Contribution

A novel view adaptation scheme integrated into an RNN with LSTM architecture that learns to select optimal viewpoints end-to-end for better action recognition.

Findings

Achieves more consistent skeleton representations across views.

Maintains action continuity without uniform frame transformation.

Outperforms state-of-the-art methods on benchmark datasets.

Abstract

Skeleton-based human action recognition has recently attracted increasing attention due to the popularity of 3D skeleton data. One main challenge lies in the large view variations in captured human actions. We propose a novel view adaptation scheme to automatically regulate observation viewpoints during the occurrence of an action. Rather than re-positioning the skeletons based on a human defined prior criterion, we design a view adaptive recurrent neural network (RNN) with LSTM architecture, which enables the network itself to adapt to the most suitable observation viewpoints from end to end. Extensive experiment analyses show that the proposed view adaptive RNN model strives to (1) transform the skeletons of various views to much more consistent viewpoints and (2) maintain the continuity of the action rather than transforming every frame to the same position with the same body orientation. Our model achieves significant improvement over the state-of-the-art approaches on three benchmark datasets.