Semantic Human Parsing via Scalable Semantic Transfer over Multiple Label Domains

被引:3
|
作者
Yang, Jie [1 ]
Wang, Chaoqun [1 ]
Li, Zhen [1 ]
Wang, Junle [2 ]
Zhang, Ruimao [1 ]
机构
[1] Chinese Univ Hong Kong, Shenzhen, Peoples R China
[2] Tencent, Shenzhen, Peoples R China
基金
中国国家自然科学基金;
关键词
D O I
10.1109/CVPR52729.2023.01861
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
This paper presents Scalable Semantic Transfer (SST), a novel training paradigm, to explore how to leverage the mutual benefits of the data from different label domains (i.e. various levels of label granularity) to train a powerful human parsing network. In practice, two common application scenarios are addressed, termed universal parsing and dedicated parsing, where the former aims to learn homogeneous human representations from multiple label domains and switch predictions by only using different segmentation heads, and the latter aims to learn a specific domain prediction while distilling the semantic knowledge from other domains. The proposed SST has the following appealing benefits: (1) it can capably serve as an effective training scheme to embed semantic associations of human body parts from multiple label domains into the human representation learning process; (2) it is an extensible semantic transfer framework without predetermining the overall relations of multiple label domains, which allows continuously adding human parsing datasets to promote the training. (3) the relevant modules are only used for auxiliary training and can be removed during inference, eliminating the extra reasoning cost. Experimental results demonstrate SST can effectively achieve promising universal human parsing performance as well as impressive improvements compared to its counterparts on three human parsing benchmarks (i.e., PASCAL-Person-Part, ATR, and CIHP). Code is available at https://github.com/yangjie-cv/SST.
引用
收藏
页码:19424 / 19433
页数:10
相关论文
共 50 条
  • [41] Hierarchical Human Semantic Parsing With Comprehensive Part-Relation Modeling
    Wang, Wenguan
    Zhou, Tianfei
    Qi, Siyuan
    Shen, Jianbing
    Zhu, Song-Chun
    [J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2022, 44 (07) : 3508 - 3522
  • [42] Pose-Guided Hierarchical Semantic Decomposition and Composition for Human Parsing
    Yang, Beibei
    Yu, Changqian
    Yu, Jin-Gang
    Gao, Changxin
    Sang, Nong
    [J]. IEEE TRANSACTIONS ON CYBERNETICS, 2023, 53 (03) : 1641 - 1652
  • [43] Face shape transfer via semantic warping
    Zonglin Li
    Xiaoqian Lv
    Wei Yu
    Qinglin Liu
    Jingbo Lin
    Shengping Zhang
    [J]. Visual Intelligence, 2 (1):
  • [44] A latent variable model of synchronous syntactic-semantic parsing for multiple languages
    Univ Geneva, Dept Computer Sci, Switzerland
    不详
    [J]. CoNLL- 2009: Shared Task - Proc. Thirteenth Conf. Comput. Natural Lang. Learn., CoNLL: Shared Task, 2009, (37-42):
  • [45] Automatic Semantic Parsing of the Ground Plane in Scenarios Recorded With Multiple Moving Cameras
    Lopez-Cifuentes, Alejandro
    Escudero-Vinolo, Marcos
    Bescos, Jesus
    [J]. IEEE SIGNAL PROCESSING LETTERS, 2018, 25 (10) : 1495 - 1499
  • [46] Towards Transparent Interactive Semantic Parsing via Step-by-Step Correction
    Mo, Lingbo
    Lewis, Ashley
    Sun, Huan
    White, Michael
    [J]. FINDINGS OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS (ACL 2022), 2022, : 322 - 342
  • [47] Multiple-Layer Classifier with Label Correction for Semantic Segmentation
    Ferariu, Lavinia
    Caraiman, Simona
    [J]. 2018 22ND INTERNATIONAL CONFERENCE ON SYSTEM THEORY, CONTROL AND COMPUTING (ICSTCC), 2018, : 703 - 708
  • [48] Learning Semantic Segmentation from Multiple Datasets with Label Shifts
    Kim, Dongwan
    Tsai, Yi-Hsuan
    Suh, Yumin
    Faraki, Masoud
    Garg, Sparsh
    Chandraker, Manmohan
    Han, Bohyung
    [J]. COMPUTER VISION - ECCV 2022, PT XXVIII, 2022, 13688 : 20 - 36
  • [49] Towards Collaborative Neural-Symbolic Graph Semantic Parsing via Uncertainty
    Lin, Zi
    Liu, Jeremiah
    Shang, Jingbo
    [J]. FINDINGS OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS (ACL 2022), 2022, : 4160 - 4173
  • [50] Weakly-Supervised Image Parsing via Constructing Semantic Graphs and Hypergraphs
    Xie, Wenxuan
    Peng, Yuxin
    Xiao, Jianguo
    [J]. PROCEEDINGS OF THE 2014 ACM CONFERENCE ON MULTIMEDIA (MM'14), 2014, : 277 - 286