Densely Semantically Aligned Person Re-Identification
CVPR, Volume abs/1812.08967, 2019, Pages 667-676.
We propose a densely semantically aligned person re-identification (re-ID) framework. It fundamentally addresses the body misalignment problem caused by pose/viewpoint variations, imperfect person detection, occlusion,etc.. By leveraging the estimation of the dense semantics of a person image, we construct a set of densely semantically al...More
PPT (Upload PPT)