GaitCrafter: Diffusion Model for Biometric Preserving Gait Synthesis

TL;DR

GaitCrafter introduces a diffusion-based framework for synthesizing realistic, controllable, and privacy-preserving gait sequences from silhouettes, enhancing gait recognition performance and enabling the creation of novel identities.

Contribution

The paper presents a novel diffusion model trained on gait silhouettes for realistic, controllable gait synthesis and introduces a method to generate synthetic identities for privacy preservation.

Findings

Synthetic gait data improves recognition accuracy.

Controllable generation allows conditioning on clothing, objects, and view angle.

Synthetic identities maintain gait consistency and diversity.

Abstract

Gait recognition is a valuable biometric task that enables the identification of individuals from a distance based on their walking patterns. However, it remains limited by the lack of large-scale labeled datasets and the difficulty of collecting diverse gait samples for each individual while preserving privacy. To address these challenges, we propose GaitCrafter, a diffusion-based framework for synthesizing realistic gait sequences in the silhouette domain. Unlike prior works that rely on simulated environments or alternative generative models, GaitCrafter trains a video diffusion model from scratch, exclusively on gait silhouette data. Our approach enables the generation of temporally consistent and identity-preserving gait sequences. Moreover, the generation process is controllable-allowing conditioning on various covariates such as clothing, carried objects, and view angle. We show that incorporating synthetic samples generated by GaitCrafter into the gait recognition pipeline leads to improved performance, especially under challenging conditions. Additionally, we introduce a mechanism to generate novel identities-synthetic individuals not present in the original dataset-by interpolating identity embeddings. These novel identities exhibit unique, consistent gait patterns and are useful for training models while maintaining privacy of real subjects. Overall, our work takes an important step toward leveraging diffusion models for high-quality, controllable, and privacy-aware gait data generation.