Exploiting the ConvLSTM: Human Action Recognition using Raw Depth Video-Based Recurrent Neural Networks

TL;DR

This paper introduces and compares ConvLSTM-based neural networks for human action recognition in raw depth videos, demonstrating that the stateful mode improves accuracy and efficiency over the stateless mode.

Contribution

The study proposes two ConvLSTM architectures for HAR, highlighting the benefits of the stateful mode in improving recognition accuracy and computational efficiency.

Findings

Stateful ConvLSTM improves recognition accuracy over stateless.

Proposed models achieve competitive accuracy with lower computational cost.

Stateful mode significantly enhances performance in video-based HAR.

Abstract

As in many other different fields, deep learning has become the main approach in most computer vision applications, such as scene understanding, object recognition, computer-human interaction or human action recognition (HAR). Research efforts within HAR have mainly focused on how to efficiently extract and process both spatial and temporal dependencies of video sequences. In this paper, we propose and compare, two neural networks based on the convolutional long short-term memory unit, namely ConvLSTM, with differences in the architecture and the long-term learning strategy. The former uses a video-length adaptive input data generator (\emph{stateless}) whereas the latter explores the \emph{stateful} ability of general recurrent neural networks but applied in the particular case of HAR. This stateful property allows the model to accumulate discriminative patterns from previous frames without compromising computer memory. Experimental results on the large-scale NTU RGB+D dataset show that the proposed models achieve competitive recognition accuracies with lower computational cost compared with state-of-the-art methods and prove that, in the particular case of videos, the rarely-used stateful mode of recurrent neural networks significantly improves the accuracy obtained with the standard mode. The recognition accuracies obtained are 75.26\% (CS) and 75.45\% (CV) for the stateless model, with an average time consumption per video of 0.21 s, and 80.43\% (CS) and 79.91\%(CV) with 0.89 s for the stateful version.