Gait Recognition with Mask-based Regularization

TL;DR

This paper introduces ReverseMask, a novel mask-based regularization technique for gait recognition that improves model generalization and outperforms state-of-the-art methods on popular datasets.

Contribution

The paper proposes a new regularization method, ReverseMask, with an Inception-like block to enhance gait recognition models' discriminative ability and generalization.

Findings

ReverseMask improves baseline accuracy.

The Inception-like block outperforms existing methods.

State-of-the-art results on CASIA-B and OUMVLP datasets.

Abstract

Most gait recognition methods exploit spatial-temporal representations from static appearances and dynamic walking patterns. However, we observe that many part-based methods neglect representations at boundaries. In addition, the phenomenon of overfitting on training data is relatively common in gait recognition, which is perhaps due to insufficient data and low-informative gait silhouettes. Motivated by these observations, we propose a novel mask-based regularization method named ReverseMask. By injecting perturbation on the feature map, the proposed regularization method helps convolutional architecture learn the discriminative representations and enhances generalization. Also, we design an Inception-like ReverseMask Block, which has three branches composed of a global branch, a feature dropping branch, and a feature scaling branch. Precisely, the dropping branch can extract fine-grained representations when partial activations are zero-outed. Meanwhile, the scaling branch randomly scales the feature map, keeping structural information of activations and preventing overfitting. The plug-and-play Inception-like ReverseMask block is simple and effective to generalize networks, and it also improves the performance of many state-of-the-art methods. Extensive experiments demonstrate that the ReverseMask regularization help baseline achieves higher accuracy and better generalization. Moreover, the baseline with Inception-like Block significantly outperforms state-of-the-art methods on the two most popular datasets, CASIA-B and OUMVLP. The source code will be released.