UniPAR: A Unified Framework for Pedestrian Attribute Recognition

TL;DR

UniPAR introduces a unified Transformer-based framework for pedestrian attribute recognition that handles multiple datasets and modalities, improving cross-domain robustness and performance in challenging environments.

Contribution

The paper presents UniPAR, a novel unified model with a phased fusion encoder and dynamic classification head for diverse pedestrian attribute recognition tasks.

Findings

Achieves comparable performance to state-of-the-art methods on benchmark datasets.

Enhances cross-domain generalization through multi-dataset joint training.

Improves robustness in low light and motion blur conditions.

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