Title:	Measuring Generalisation to Unseen Viewpoints, Articulations, Shapes and Objects for 3D Hand Pose Estimation Under Hand-Object Interaction
Authors:	Armagan, Anil Garcia-Hernando, Guillermo Baek, Seungryul Hampali, Shreyas Rad, Mahdi Zhang, Zhaohui Xie, Shipeng Chen, MingXiu Zhang, Boshen Xiong, Fu Yang, Xiao Cao, Zhiguo Yuan, Junsong Ren, Pengfei Huang, Weiting Sun, Haifeng Hrúz, Marek Kanis, Jakub Krňoul, Zdeněk Wan, Qingfu Li, Shile Yang, Linlin Lee, Dongheui Yao, Angela Zhou, Weiguo Mei, Sijia Liu, Yunhui Spurr, Adrian Iqbal, Umar Molchanov, Pavlo Weinzaepfel, Philippe Brégier, Romain Rogez, Grégory Lepetit, Vincent Kim, Tae-Kyun
Citation:	ARMAGAN, A., GARCIA-HERNANDO, G., BAEK, S., HAMPALI, S., RAD, M., ZHANG, Z., XIE, S., CHEN, M., ZHANG, B., XIONG, F., YANG, X., CAO, Z., YUAN, J., REN, P., HUANG, W., SUN, H., HRÚZ, M., KANIS, J., KRŇOUL, Z., WAN, Q., LI, S., YANG, L., LEE, D., YAO, A., ZHOU, W., MEI, S., LIU, Y., SPURR, A., IQBAL, U., MOLCHANOV, P., WEINZAEPFEL, P., BRÉGIER, R., ROGEZ, G., LEPETIT, V., KIM, T. Measuring Generalisation to Unseen Viewpoints, Articulations, Shapes and Objects for 3D Hand Pose Estimation Under Hand-Object Interaction. In: Computer Vision - ECCV 2020, 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part XXIII. Cham: Springer, 2020. s. 85-101. ISBN 978-3-030-58591-4, ISSN 0302-9743.
Issue Date:	2020
Publisher:	Springer
Document type:	konferenční příspěvek conferenceObject
URI:	2-s2.0-85097407772 http://hdl.handle.net/11025/42724
ISBN:	978-3-030-58591-4
ISSN:	0302-9743
Keywords in different language:	Hand Pose Estiamtion
Abstract in different language:	We study how well different types of approaches generalise in the task of 3D hand pose estimation under single hand scenarios and hand-object interaction. We show that the accuracy of state-of-the-art methods can drop, and that they fail mostly on poses absent from the training set. Unfortunately, since the space of hand poses is highly dimensional, it is inherently not feasible to cover the whole space densely, despite recent efforts in collecting large-scale training datasets. This sampling problem is even more severe when hands are interacting with objects and/or inputs are RGB rather than depth images, as RGB images also vary with lighting conditions and colors. To address these issues, we designed a public challenge (HANDS’19) to evaluate the abilities of current 3D hand pose estimators (HPEs) to interpolate and extrapolate the poses of a training set. More exactly, HANDS’19 is designed (a) to evaluate the influence of both depth and color modalities on 3D hand pose estimation, under the presence or absence of objects; (b) to assess the generalisation abilities w.r.t. four main axes: shapes, articulations, viewpoints, and objects; (c) to explore the use of a synthetic hand models to fill the gaps of current datasets. Through the challenge, the overall accuracy has dramatically improved over the baseline, especially on extrapolation tasks, from 27 mm to 13 mm mean joint error. Our analyses highlight the impacts of: Data pre-processing, ensemble approaches, the use of a parametric 3D hand model (MANO), and different HPE methods/backbones.
Rights:	Plný text není přístupný. © Springer
Appears in Collections:	Konferenční příspěvky / Conference papers (NTIS) Konferenční příspěvky / Conference Papers (KKY) OBD

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.