UniPAR: A Unified Framework for Pedestrian Attribute Recognition

2026-03-05

AI Summary

This work addresses the limited generalization and dataset-specific model training challenges in pedestrian attribute recognition caused by heterogeneous datasets—differing in modality, attribute definitions, and environmental conditions—by proposing UniPAR, the first unified Transformer framework capable of cross-modal (RGB images, videos, event streams) and cross-dataset learning. UniPAR leverages a unified data scheduling strategy, dynamic classification heads, and a staged fusion encoder to achieve deep alignment between visual features and textual attribute queries. Evaluated on benchmarks including MSP60K, DukeMTMC, and EventPAR, UniPAR matches the performance of specialized state-of-the-art methods while significantly enhancing robustness and cross-domain generalization under extreme conditions such as low light and motion blur through joint multi-dataset training.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Multimodal Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs

Abstract

Pedestrian Attribute Recognition is a foundational computer vision task that provides essential support for downstream applications, including person retrieval in video surveillance and intelligent retail analytics. However, existing research is frequently constrained by the ``one-model-per-dataset"paradigm and struggles to handle significant discrepancies across domains in terms of modalities, attribute definitions, and environmental scenarios. To address these challenges, we propose UniPAR, a unified Transformer-based framework for PAR. By incorporating a unified data scheduling strategy and a dynamic classification head, UniPAR enables a single model to simultaneously process diverse datasets from heterogeneous modalities, including RGB images, video sequences, and event streams. We also introduce an innovative phased fusion encoder that explicitly aligns visual features with textual attribute queries through a late deep fusion strategy. Experimental results on the widely used benchmark datasets, including MSP60K, DukeMTMC, and EventPAR, demonstrate that UniPAR achieves performance comparable to specialized SOTA methods. Furthermore, multi-dataset joint training significantly enhances the model's cross-domain generalization and recognition robustness in extreme environments characterized by low light and motion blur. The source code of this paper will be released on https://github.com/Event-AHU/OpenPAR

Problem

• Pedestrian Attribute Recognition
• domain discrepancy
• multi-modality
• unified framework
• cross-domain generalization

Innovation

• Unified Framework
• Transformer-based PAR
• Multi-modality Fusion
• Dynamic Classification Head
• Cross-domain Generalization