Database-Agnostic Gait Enrollment using SetTransformers

2025-05-05

AI Summary

This work addresses the open-set gait registration problem—determining whether a novel gait sample belongs to a known identity or an unknown individual—by proposing the first dataset- and recognition-architecture-agnostic framework. Methodologically, it decouples registration from recognition: SetTransformer models skeleton sequences to produce embeddings compatible with diverse gait representation methods (e.g., GaitGraph, GaitFormer, GaitPT); supports dynamic gallery sizes and evolving identity distributions; and requires no task-specific thresholds or model retraining. Evaluated on CASIA-B and PsyMo, the framework demonstrates strong cross-dataset generalization and scalability, achieving significantly higher registration accuracy than conventional threshold-based baselines. To foster reproducibility and community advancement, the authors will release the code, pre-trained models, and a standardized open-set evaluation protocol.

Technology Category

• Computer Vision: Biometrics, Face, Gesture & Pose
• Machine Learning: Calibration & Uncertainty Quantification
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Security and Privacy: Data transparency and provenance
• User Modeling, Personalization and Recommendation: Accountability, Transparency, and Ethics for personalization
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Gait recognition has emerged as a powerful tool for unobtrusive and long-range identity analysis, with growing relevance in surveillance and monitoring applications. Although recent advances in deep learning and large-scale datasets have enabled highly accurate recognition under closed-set conditions, real-world deployment demands open-set gait enrollment, which means determining whether a new gait sample corresponds to a known identity or represents a previously unseen individual. In this work, we introduce a transformer-based framework for open-set gait enrollment that is both dataset-agnostic and recognition-architecture-agnostic. Our method leverages a SetTransformer to make enrollment decisions based on the embedding of a probe sample and a context set drawn from the gallery, without requiring task-specific thresholds or retraining for new environments. By decoupling enrollment from the main recognition pipeline, our model is generalized across different datasets, gallery sizes, and identity distributions. We propose an evaluation protocol that uses existing datasets in different ratios of identities and walks per identity. We instantiate our method using skeleton-based gait representations and evaluate it on two benchmark datasets (CASIA-B and PsyMo), using embeddings from three state-of-the-art recognition models (GaitGraph, GaitFormer, and GaitPT). We show that our method is flexible, is able to accurately perform enrollment in different scenarios, and scales better with data compared to traditional approaches. We will make the code and dataset scenarios publicly available.

Problem

• Open-set gait enrollment for identity analysis
• Dataset-agnostic and architecture-agnostic framework
• Flexible enrollment across diverse scenarios

Innovation

• Transformer-based open-set gait enrollment framework
• SetTransformer for dataset-agnostic enrollment decisions
• Decouples enrollment from recognition pipeline