# CDPM: Convolutional Deformable Part Models for Semantically Aligned Person Re-identification

## TL;DR

This paper introduces CDPM, a novel model that improves person re-identification by sequentially aligning body parts vertically and horizontally, reducing misalignment issues caused by detection errors.

## Contribution

The paper proposes a two-step orthogonal approach for part alignment in person re-identification, significantly enhancing accuracy without external information.

## Key findings

- Achieves state-of-the-art results on three large datasets
- Effectively reduces misalignment caused by detection errors
- Demonstrates robustness of the sequential alignment approach

## Abstract

Part-level representations are essential for robust person re-identification. However, common errors that arise during pedestrian detection frequently result in severe misalignment problems for body parts, which degrade the quality of part representations. Accordingly, to deal with this problem, we propose a novel model named Convolutional Deformable Part Models (CDPM). CDPM works by decoupling the complex part alignment procedure into two easier steps: first, a vertical alignment step detects each body part in the vertical direction, with the help of a multi-task learning model; second, a horizontal refinement step based on attention suppresses the background information around each detected body part. Since these two steps are performed orthogonally and sequentially, the difficulty of part alignment is significantly reduced. In the testing stage, CDPM is able to accurately align flexible body parts without any need for outside information. Extensive experimental results demonstrate the effectiveness of the proposed CDPM for part alignment. Most impressively, CDPM achieves state-of-the-art performance on three large-scale datasets: Market-1501, DukeMTMC-ReID,and CUHK03.