Publications

Detailed Information

Learning spectral transform for 3D human motion prediction

Cited 0 time in Web of Science Cited 1 time in Scopus
Authors

Kim, Boeun; Choi, Jin Young

Issue Date
2022-10
Publisher
Academic Press
Citation
Computer Vision and Image Understanding, Vol.223, p. 103548
Abstract
© 2022 Elsevier Inc.In existing motion prediction methods that use graph convolution networks, motion sequences are transformed to a spectral domain, and future motions are predicted through graph spectral filtering for the transformed spectral sequences. However, because the conventional spectral transform method uses a predetermined spectral basis, the motion prediction does not work well for aperiodic or complicated motions. To overcome this problem, we propose a method to learn spectral domain transforms from motion sequences in the training dataset. To this end, two methods are attempted: one for learning the frequency of each spectral basis, and another for learning the values of the basis function directly. Through experiments on representative 3D human motion benchmarks, H3.6M and CMU Mocap, we demonstrate that both of the proposed methods consistently outperform the baseline method. In particular, the method of directly learning the basis function outperforms the state-of-the-art methods. We also demonstrate that the proposed method yields realistic predictions, even for aperiodic and complicated action categories.
ISSN
1077-3142
URI
https://hdl.handle.net/10371/186255
DOI
https://doi.org/10.1016/j.cviu.2022.103548
Files in This Item:
There are no files associated with this item.
Appears in Collections:

Altmetrics

Item View & Download Count

  • mendeley

Items in S-Space are protected by copyright, with all rights reserved, unless otherwise indicated.

Share