Human-Object Interaction Detection via Disentangled Transformer

doi:10.1109/CVPR52688.2022.01896

	Human-Object Interaction Detection via Disentangled Transformer
	Zhou, Desen 1; Liu, Zhichao1,2 ; Wang, Jian 1; Wang, Leshan1,2 ; Hu, Tao 1; Ding, Errui 1; Wang, Jingdong 1
	2022
会议录名称	2022 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2022)
ISSN	1063-6919
发表状态	已发表
DOI	10.1109/CVPR52688.2022.01896
摘要	Human-Object Interaction Detection tackles the problem of joint localization and classification of human object interactions. Existing HOI transformers either adopt a single decoder for triplet prediction, or utilize two parallel decoders to detect individual objects and interactions separately, and compose triplets by a matching process. In contrast, we decouple the triplet prediction into human-object pair detection and interaction classification. Our main motivation is that detecting the human-object instances and classifying interactions accurately needs to learn representations that focus on different regions. To this end, we present Disentangled Transformer, where both encoder and decoder are disentangled to facilitate learning of two subtasks. To associate the predictions of disentangled decoders, we first generate a unified representation for HOI triplets with a base decoder, and then utilize it as input feature of each disentangled decoder. Extensive experiments show that our method outperforms prior work on two public HOI benchmarks by a sizeable margin. Code will be available.
会议名称	IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
出版地	10662 LOS VAQUEROS CIRCLE, PO BOX 3014, LOS ALAMITOS, CA 90720-1264 USA
会议地点	null,New Orleans,LA
会议日期	JUN 18-24, 2022
URL	查看原文
收录类别	CPCI-S
语种	英语
WOS研究方向	Computer Science ; Imaging Science & Photographic Technology
WOS类目	Computer Science, Artificial Intelligence ; Imaging Science & Photographic Technology
WOS记录号	WOS:000870783005037
出版者	IEEE COMPUTER SOC
来源库	IEEE
引用统计	正在获取...
文献类型	会议论文
条目标识符	https://kms.shanghaitech.edu.cn/handle/2MSLDSTB/266587
专题	信息科学与技术学院_硕士生
通讯作者	Zhou, Desen
作者单位	1.Baidu Inc, Dept Comp Vis Technol Vis, Beijing, Peoples R China 2.ShanghaiTech Univ, Shanghai, Peoples R China
推荐引用方式 GB/T 7714	Zhou, Desen,Liu, Zhichao,Wang, Jian,et al. Human-Object Interaction Detection via Disentangled Transformer[C]. 10662 LOS VAQUEROS CIRCLE, PO BOX 3014, LOS ALAMITOS, CA 90720-1264 USA:IEEE COMPUTER SOC,2022.