Watch Where You Head: A View-biased Domain Gap in Gait Recognition and Unsupervised Adaptation

TL;DR

This paper identifies a view bias problem in gait recognition domain adaptation and proposes GOUDA, a novel method using triplet selection and curriculum learning to improve model generalization across unseen viewing angles.

Contribution

The paper introduces GOUDA, a new unsupervised domain adaptation technique specifically addressing view bias in gait recognition models, with extensive experimental validation.

Findings

GOUDA outperforms prior UDA methods on multiple datasets.

View bias significantly affects gait recognition performance.

The proposed method effectively reduces view bias in various backbones.

Abstract

Gait Recognition is a computer vision task aiming to identify people by their walking patterns. Although existing methods often show high performance on specific datasets, they lack the ability to generalize to unseen scenarios. Unsupervised Domain Adaptation (UDA) tries to adapt a model, pre-trained in a supervised manner on a source domain, to an unlabelled target domain. There are only a few works on UDA for gait recognition proposing solutions to limited scenarios. In this paper, we reveal a fundamental phenomenon in adaptation of gait recognition models, caused by the bias in the target domain to viewing angle or walking direction. We then suggest a remedy to reduce this bias with a novel triplet selection strategy combined with curriculum learning. To this end, we present Gait Orientation-based method for Unsupervised Domain Adaptation (GOUDA). We provide extensive experiments on four widely-used gait datasets, CASIA-B, OU-MVLP, GREW, and Gait3D, and on three backbones, GaitSet, GaitPart, and GaitGL, justifying the view bias and showing the superiority of our proposed method over prior UDA works.