Silhouette based View embeddings for Gait Recognition under Multiple Views

TL;DR

This paper introduces a novel framework for gait recognition across multiple views by embedding view information into existing CNN architectures using a selective projection layer, improving recognition accuracy.

Contribution

It proposes a simple yet effective view embedding method for gait recognition that does not require view estimation or training separate models.

Findings

Significant improvement in recognition accuracy on public datasets.

Effective integration of view information into existing CNN models.

No need for explicit view angle estimation or view-specific training.

Abstract

Gait recognition under multiple views is an important computer vision and pattern recognition task. In the emerging convolutional neural network based approaches, the information of view angle is ignored to some extent. Instead of direct view estimation and training view-specific recognition models, we propose a compatible framework that can embed view information into existing architectures of gait recognition. The embedding is simply achieved by a selective projection layer. Experimental results on two large public datasets show that the proposed framework is very effective.