… We refer to such a Gaussian distribution representing each local patch as a patch Gaussian.
… We refer to this Gaussian distribution as a region Gaussian. In both steps, we embed the …

… Person Re-Identification Person re-identification has gained increasing attention from both
… Then we refer to these similarity values and use an attention mechanism to adaptively …

… a global feature of a person image itself, gives robust … persons having similar attributes in
corresponding parts. To address this issue, we propose a new relation network for person reID …

… success in person re-identification by learning powerful person representations [16, 17].
Deep learning based methods typically work on extracting a global feature of a person image [33…

… Yet in this paper, we concentrate on a high quality pedestrian descriptor through better …
(For interpretation of the references to color in this figure legend, the reader is referred to the …

… descriptor; feature comparison assigns the ID with respect to the re-projected DB. (For
interpretation of the references … achieving short-term person re-identification into indoor scenarios. …

… The two gaps hinder the further development of person re-identification. To address this …
It uses a cloth-template as a reference to retrieve the target person. This setting reduces the …

… Wu, “Efficient online local metric adaptation via negative samples for person re-identification,…
Miao, “Multiple metric learning based on bar-shape descriptor for person reidentification,” …

… CONCLUSION This paper explicitly addresses the pedestrian misalignment problem in
person re-identification. We propose the pose invariant embedding (PIE) as pedestrian descriptor…

… , a histogram descriptor is generated and concatenated with the full body CNN feature. Wei
et al. [17] design a robust and discriminative Global-Local-Alignment Descriptor (GLAD) to …