Metric Learning to improve the persistent homology-based gait recognition

Abstract

Gait recognition is an important biometric technique for video surveillance tasks, due to the advantage of using it at distance. In this paper, we present a persistent homology-based method to extract topological features from the body silhouettes of a gait sequence. It has been used before in several papers for the second author for human identification, gender classification, carried object detection and monitoring human activities at distance. As the previous work, we apply persistent homology to extract topological features from the lowest fourth part of the body silhouette to decrease the negative effects of variations unrelated to the gait in the upper body part. The novelty of this paper is the introduction of the use of a metric learning to learn a Mahalanobis distance metric to robust gait recognition, where we use Linear Discriminant Analysis. This learned metric enforces objects for the same class to be closer while objects from different classes are pulled apart. We evaluate our approach using the CASIA-B dataset and we show the effectiveness of the methods proposed compared with other state-of-the-art methods.

Keywords

• TDA
• gait recognition
• persistent homology
• linear discriminant analysis
• metric learning.