Search papers, labs, and topics across Lattice.
This paper introduces UniGeo, a Universal Monocular 3D-Enhanced ReID framework that addresses the challenges of pedestrian identity retrieval in complex real-world scenarios by integrating 3D geometric information. By employing a Consistency-Aware Reliability Gate and Dual-Stream Residual Fusion, the method effectively decouples geometric extraction from dynamic utilization, allowing for robust handling of severe occlusions and clothing changes. Experimental results demonstrate that UniGeo significantly enhances performance in structure-sensitive conditions while maintaining competitive results in cleaner environments.
Monocular 3D priors can enhance person re-identification without succumbing to noise, achieving better results in challenging visual conditions.
Universal person re-identification (ReID) aims to retrieve pedestrian identities across diverse real-world scenarios, including severe occlusions, clothing changes, and cross-modality shifts, within a unified model. However, existing 2D representations fundamentally struggle with spatial ambiguities due to a lack of depth and topological awareness, while naively introducing monocular 3D priors often causes severe negative transfer due to geometric estimation noise under extreme visual degradation. To safely harness the clothing-invariant and canonical structural properties of 3D geometry, we propose UniGeo, a Universal Monocular 3D-Enhanced ReID framework driven by a Consistency-Aware Reliability Gate and Dual-Stream Residual Fusion. Specifically, the processing of 3D information is strategically decoupled into geometric extraction and dynamic utilization. To provide pure structural compensation, we project monocular 3D parameters into kinematic joint representations, explicitly capturing instance-level geometric topology to resolve appearance-based ambiguities. To robustly incorporate these cues without perturbing the reliable 2D feature space, we isolate the 3D prior as a late-stage structural residual; modulated by the consistency-aware gate, this mechanism adaptively filters geometric noise and enables controlled fallback to the pure 2D baseline. Extensive experiments show that our method improves challenging, structure-sensitive scenarios while preserving competitive performance on clean domains. Code is available at https://github.com/BohanSu/UniGeo.
