|
作者 | Jiayu, Zou1,2 ; Jie, Qin1,2 ; Zhen, Zhang1,2 ; Xingang, Wang1
|
出版日期 | 2022-06
|
会议日期 | 2022-07
|
会议地点 | Xi'an, China
|
页码 | 300-306
|
英文摘要 | Human pose estimation is broadly used in action recognition, Re-Identity, and multi-object tracking. Recently deep
convolutional neural networks have demonstrated their great power in human pose estimation. However, CNN-based methods are limited by the constrained receptive field that has poor performance in modeling global relationships of different body parts. In this paper, we propose a novel multi-granularity feature interaction network for human pose estimation (MITPose), which exploits the multi-granularity feature interaction in global-local level features, multi-scale features, and locality features. Our MITPose can efficiently leverage the long-range representation ability of transformer net and inductive locality of convolution net to obtain the comprehensive information for key point localization and relationship modeling. Extensive experiments illustrate that our proposed MITPose achieves state-of-the-art performance on the public COCO dataset. |
语种 | 英语
|
源URL | [http://ir.ia.ac.cn/handle/173211/51537]  |
专题 | 精密感知与控制研究中心_精密感知与控制
|
通讯作者 | Xingang, Wang |
作者单位 | 1.中国科学院大学自动化研究所 2.中国科学院大学
|
推荐引用方式 GB/T 7714 |
Jiayu, Zou,Jie, Qin,Zhen, Zhang,et al. MITPose: Multi-Granularity Feature Interaction for Human Pose Estimation[C]. 见:. Xi'an, China. 2022-07.
|