PoSeg: Pose-Aware Refinement Network for Human Instance Segmentation

Human instance segmentation is a core problem for human-centric scene understanding and segmenting human instances poses a unique challenge to vision systems due to large intra-class variations in both appearance and shape, and complicated occlusion patterns. In this paper, we propose a new pose-awa...

Полное описание

Библиографические подробности
Главные авторы: Desen Zhou, Qian He
Формат: Статья
Язык:English
Опубликовано: IEEE 2020-01-01
Серии:IEEE Access
Предметы:
Online-ссылка:https://ieeexplore.ieee.org/document/8962018/