Research on Multi-level Attention-based Human Pose Estimation

Detecting human key points from a single image is very challenging due to occlusion, blurring, illumination and scale changes. In this paper, this problem is addressed by designing an effective network structure. Since global and local information plays an important role in reasoning about human bod...

Full description

Saved in:
Bibliographic Details
Published in2022 2nd International Conference on Algorithms, High Performance Computing and Artificial Intelligence (AHPCAI) pp. 592 - 595
Main Authors Gao, Jun, Huang, Hua, Li, QiShen, Xia, ShaoQi
Format Conference Proceeding
LanguageEnglish
Published IEEE 21.10.2022
Subjects
attention mechanism
Cognition
Computational modeling
Feature extraction
High performance computing
Human Pose Estimation
Keypoint detection
Lighting
Pose estimation
Sun
Online AccessGet full text

Cover

More Information
Summary:Detecting human key points from a single image is very challenging due to occlusion, blurring, illumination and scale changes. In this paper, this problem is addressed by designing an effective network structure. Since global and local information plays an important role in reasoning about human body structure and invisible keypoints, Multi-level Attention Network (MAN) is proposed. First, compared with traditional multi-resolution networks, it enables multi-resolution feature maps with greater information variance by generating them directly from the highest resolution feature map, which in turn increases the abundance of feature information after final fusion. Secondly, it effectively integrates global and local information in different resolution feature maps through the Feature Alignment Attention Block(FAAB), and intensifies them in a targeted manner. On the COCO dataset, with HRNet (Sun K. et al [1]) as the baseline network, HRNet of inserted MAN improves 1.1-2.3 AP points over the baseline network.
DOI:10.1109/AHPCAI57455.2022.10087380