Resolution as a Direction: Vector-Panning Feature Alignment for Cross-Resolution Re-Identification

Cross-resolution person re-identification (CR-ReID) remains challenging in practical surveillance, where camera quality and capture distance lead to substantial resolution gaps between low-resolution (LR) queries and high-resolution (HR) gallery images. Prior approaches commonly rely on super-resolution (SR) or resolution-invariant representation learning, which often increases system complexity and may not directly address the feature mismatch induced by resolution degradation. In this work, we report a new empirical finding from a dedicated analysis in which identity-specific variation is averaged out: the HR--LR feature discrepancy produced by standard ReID backbones exhibits a consistent, resolution-related semantic direction in the embedding space. We further support this observation with statistical analyses based on Canonical Correlation Analysis (CCA) and Pearson correlation analysis. Motivated by this finding, we propose Vector Panning Feature Alignment (VPFA), a lightweight post-hoc module that learns to pan LR features along the learned resolution direction to obtain pseudo-HR representations. VPFA operates after feature extraction and can be integrated into existing ReID systems with negligible overhead. Extensive experiments on multiple CR-ReID benchmarks show that VPFA achieves state-of-the-art performance while improving efficiency compared to SR-based or jointly trained alternatives. Code is available at \url{https://github.com/ashmentlzw/VPFA}

Person re-identification aims to match images of the same person across different cameras. In real-world camera systems, this is difficult because one camera may capture a clear, high-resolution image while another captures a small or blurry low-resolution image. Existing methods often try to improve the low-resolution image itself, but this can be slow and may not always help the final matching task. This paper studies the problem from a different angle. We find that low-resolution images tend to affect the model’s internal representation in a consistent way, almost like shifting it in a particular direction. Based on this observation, we propose a lightweight method that adjusts the representation of a low-resolution image so that it becomes more like the representation of a high-resolution image of the same person. Our method works after feature extraction, so it can be added to existing person re-identification systems without retraining the whole model. Experiments on several benchmarks show that it improves matching accuracy while adding very little computational cost. This makes it practical for real-world settings where images often have different resolutions.