HOME: 3D Human–Object Mesh Topology-Enhanced Interaction Recognition in Images

Author:

Peng Weilong,Li Cong,Tang Keke,Liu Xianyong,Fang MeieORCID

Abstract

Human–object interaction (HOI) recognition is a very challenging task due to the ambiguity brought by occlusions, viewpoints, and poses. Because of the limited interaction information in the image domain, extracting 3D features of a point cloud has been an important means to improve the recognition performance of HOI. However, the features neglect topological features of adjacent points at low level, and the deep topology relation between a human and an object at high level. In this paper, we present a 3D human–object mesh topology enhanced method (HOME) for HOI recognition in images. In the method, human–object mesh (HOM) is built by integrating the reconstructed human and object mesh from images firstly. Therefore, under the assumption that the interaction comes from the macroscopic pattern constructed by spatial position and microscopic topology of human–object, HOM is inputted into MeshCNN to extract the effective edge features by edge-based convolution from bottom to up, as the topological features that encode the invariance of the interaction relationship. At last, topological cues are fused with visual cues to enhance the recognition performance greatly. In the experiment, HOI recognition results have achieved an improvement of about 4.3% mean average precision (mAP) in the Rare cases of the HICO-DET dataset, which verifies the effectiveness of the proposed method.

Funder

National Natural Science Foundation of China

Guangdong Basic and Applied Basic Research Foundation

Publisher

MDPI AG

Subject

General Mathematics,Engineering (miscellaneous),Computer Science (miscellaneous)

Reference41 articles.

1. End-to-end concept word detection for video captioning, retrieval, and question answering;Yu;Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition,2017

2. A joint sequence fusion model for video question answering and retrieval;Yu;Proceedings of the European Conference on Computer Vision (ECCV),2018

3. Mdmmt: Multidomain multimodal transformer for video retrieval;Dzabraev;Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition,2021

4. Quo vadis, action recognition? A new model and the kinetics dataset;Carreira;Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition,2017

5. T-C3D: Temporal convolutional 3D network for real-time action recognition;Liu;Proceedings of the AAAI Conference on Artificial Intelligence,2018

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3