Fast Human Pose Estimation

被引:163
|
作者
Zhang, Feng [1 ]
Zhu, Xiatian [2 ]
Ye, Mao [1 ]
机构
[1] Univ Elect Sci & Technol China, Sch Comp Sci & Engn, Chengdu, Peoples R China
[2] Vis Semant Ltd, London, England
基金
中国国家自然科学基金;
关键词
D O I
10.1109/CVPR.2019.00363
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Existing human pose estimation approaches often only consider how to improve the model generalisation performance, but putting aside the significant efficiency problem. This leads to the development of heavy models with poor scalability and cost-effectiveness in practical use. In this work, we investigate the under-studied but practically critical pose model efficiency problem. To this end, we present a new Fast Pose Distillation (FPD) model learning strategy. Specifically, the FPD trains a lightweight pose neural network architecture capable of executing rapidly with low computational cost. It is achieved by effectively transferring the pose structure knowledge of a strong teacher network. Extensive evaluations demonstrate the advantages of our FPD method over a broad range of state-of-the-art pose estimation approaches in terms of model cost-effectiveness on two standard benchmark datasets, MPII Human Pose and Leeds Sports Pose.
引用
收藏
页码:3512 / 3521
页数:10
相关论文
共 50 条
  • [31] Fast pose estimation with parameter-sensitive hashing
    Shakhnarovich, G
    Viola, P
    Darrell, T
    [J]. NINTH IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION, VOLS I AND II, PROCEEDINGS, 2003, : 750 - 757
  • [32] Fast certifiable relative pose estimation with gravity prior*
    Garcia-Salguero, Mercedes
    Gonzalez-Jimenez, Javier
    [J]. ARTIFICIAL INTELLIGENCE, 2023, 317
  • [33] Fast Certifiable Algorithm for the Absolute Pose Estimation of a Camera
    Garcia-Salguero, Mercedes
    Dima, Elijs
    Mateus, Andrée
    Gonzalez-Jimenez, Javier
    [J]. SIAM Journal on Imaging Sciences, 2024, 17 (03): : 1415 - 1432
  • [34] Fast Head Pose Estimation using Depth Data
    Qiao, Ti-zhou
    Dai, Shu-ling
    [J]. 2013 6TH INTERNATIONAL CONGRESS ON IMAGE AND SIGNAL PROCESSING (CISP), VOLS 1-3, 2013, : 664 - 669
  • [35] Fast Relative Pose Estimation using Relative Depth
    Astermark, Jonathan
    Ding, Yaqing
    Larsson, Viktor
    Heyden, Anders
    [J]. 2024 INTERNATIONAL CONFERENCE IN 3D VISION, 3DV 2024, 2024, : 873 - 881
  • [36] ADAPTIVE FREQUENCY HOPPING POLICY FOR FAST POSE ESTIMATION
    Liang, Yuchen
    Liu, Yuehu
    [J]. 2021 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP), 2021, : 324 - 328
  • [37] Fast and Efficient Root Phenotyping via Pose Estimation
    Berrigan, Elizabeth M.
    Wang, Lin
    Carrillo, Hannah
    Echegoyen, Kimberly
    Kappes, Mikayla
    Torres, Jorge
    Ai-Perreira, Angel
    Mccoy, Erica
    Shane, Emily
    Copeland, Charles D.
    Ragel, Lauren
    Georgousakis, Charidimos
    Lee, Sanghwa
    Reynolds, Dawn
    Talgo, Avery
    Gonzalez, Juan
    Zhang, Ling
    Rajurkar, Ashish B.
    Ruiz, Michel
    Daniels, Erin
    Maree, Liezl
    Pariyar, Shree
    Busch, Wolfgang
    Pereira, Talmo D.
    [J]. PLANT PHENOMICS, 2024, 6
  • [38] Fast Iterative Five point Relative Pose Estimation
    Hedborg, Johan
    Felsberg, Michael
    [J]. 2013 IEEE WORKSHOP ON ROBOT VISION (WORV), 2013, : 60 - 67
  • [39] Human Pose Estimation in Stereo Images
    Lallemand, Joe
    Szczot, Magdalena
    Ilic, Slobodan
    [J]. ARTICULATED MOTION AND DEFORMABLE OBJECTS, AMDO 2014, 2014, 8563 : 10 - 19
  • [40] An Improved Silhouette for Human Pose Estimation
    Hawes, Anthony H.
    Iftekharuddin, Khan M.
    [J]. OPTICS AND PHOTONICS FOR INFORMATION PROCESSING XI, 2017, 10395