SAS-VPReID: A Scale-Adaptive Framework with Shape Priors for Video-based Person Re-Identification at Extreme Far Distances

TL;DR

SAS-VPReID introduces a scale-adaptive framework with shape priors that significantly improves video-based person re-identification at extreme far distances by addressing resolution, viewpoint, and appearance challenges.

Contribution

The paper presents a novel framework combining a memory-enhanced visual backbone, multi-granularity temporal modeling, and shape dynamics regularization for improved VPReID performance.

Findings

Achieved top rank on VReID-XFD benchmark.

Each module significantly enhances re-identification accuracy.

Framework effectively handles extreme far-distance challenges.

Abstract

Video-based Person Re-IDentification (VPReID) aims to retrieve the same person from videos captured by non-overlapping cameras. At extreme far distances, VPReID is highly challenging due to severe resolution degradation, drastic viewpoint variation and inevitable appearance noise. To address these issues, we propose a Scale-Adaptive framework with Shape Priors for VPReID, named SAS-VPReID. The framework is built upon three complementary modules. First, we deploy a Memory-Enhanced Visual Backbone (MEVB) to extract discriminative feature representations, which leverages the CLIP vision encoder and multi-proxy memory. Second, we propose a Multi-Granularity Temporal Modeling (MGTM) to construct sequences at multiple temporal granularities and adaptively emphasize motion cues across scales. Third, we incorporate Prior-Regularized Shape Dynamics (PRSD) to capture body structure dynamics. With these modules, our framework can obtain more discriminative feature representations. Experiments on the VReID-XFD benchmark demonstrate the effectiveness of each module and our final framework ranks the first on the VReID-XFD challenge leaderboard. The source code is available at https://github.com/YangQiWei3/SAS-VPReID.