Litcius/Paper detail

Partial Person Re-identification with Part-Part Correspondence Learning

Tianyu He, Xu Shen, Jianqiang Huang, Zhibo Chen, Xian‐Sheng Hua

202146 citationsDOI

Abstract

Driven by the success of deep learning, the last decade has seen rapid advances in person re-identification (re-ID). Nonetheless, most of approaches assume that the input is given with the fulfillment of expectations, while imperfect input remains rarely explored to date, which is a non-trivial problem since directly apply existing methods without adjustment can cause significant performance degradation. In this paper, we focus on recognizing partial (flawed) input with the assistance of proposed Part-Part Correspondence Learning (PPCL), a self-supervised learning framework that learns correspondence between image patches without any additional part-level supervision. Accordingly, we propose Part-Part Cycle (PP-Cycle) constraint and Part-Part Triplet (PP-Triplet) constraint that exploit the duality and uniqueness between corresponding image patches respectively. We verify our proposed PPCL on several partial person re-ID benchmarks. Experimental results demonstrate that our approach can surpass previous methods in terms of the standard evaluation metric.

Topics & Concepts

Constraint (computer-aided design)Identification (biology)Computer scienceMetric (unit)Artificial intelligenceFocus (optics)ExploitImperfectImage (mathematics)Duality (order theory)UniquenessDeep learningMachine learningAlgorithmPattern recognition (psychology)MathematicsEngineeringDiscrete mathematicsBiologyLinguisticsComputer securityGeometryBotanyOpticsPhilosophyPhysicsOperations managementMathematical analysisVideo Surveillance and Tracking MethodsAdvanced Neural Network ApplicationsHuman Pose and Action Recognition