Dr. BAI Junxuan, a full-time researcher at the Institute of Artificial Intelligence in Sports (IAIS), has published the latest research results in the field of motion prediction as a co-author, and the related paper has been published in the international journal Pattern Recognition, entitled “KD-Former: Kinematic and dynamic coupled transformer network for 3D human motion prediction”.
Abstract
Recent studies have made remarkable progress on 3D human motion prediction by describing motion with kinematic knowledge. However, kinematics only considers the 3D positions or rotations of human skeletons, failing to reveal the physical characteristics of human motion. Motion dynamics reflects the forces between joints, explicitly encoding the skeleton topology, whereas rarely exploited in motion prediction. In this paper, we propose the Kinematic and Dynamic coupled transFormer (KD-Former), which incorporates dynamics with kinematics, to learn powerful features for high-fidelity motion prediction. Specifically, We first formulate a reduced-order dynamic model of human body to calculate the forces of all joints. Then we construct a non-autoregressive encoder-decoder framework based on the transformer structure. The encoder involves a kinematic encoder and a dynamic encoder, which are respectively responsible for extracting the kinematic and dynamic features for given history sequences via a spatial transformer and a temporal transformer. Future query sequences are decoded in parallel in the decoder by leveraging the encoded kinematic and dynamic information of history sequences. Experiments on Human3.6M and CMU MoCap benchmarks verify the effectiveness and superiority of our method. Code will be available at: https://github.com/wslh852/KD-Former.git.
Journal Introduction
Pattern Recognition is a category B journal of artificial intelligence (CCF-B journal) in the “Recommended International Academic Journals” of the China Computer Federation. It belongs to SCI Q1 journal, and is in the 25/145 position in the field of computer science and artificial intelligence. The impact factor reached 8.0 in 2022. Pattern Recognition is a mature but exciting rapidly developing field that lays the foundation for the development of related fields such as computer vision, image processing, text and document analysis, and neural networks. It is closely related to machine learning and has been applied in rapidly emerging fields such as biometrics, bioinformatics, multimedia data analysis, and more recently, data science. Pattern Recognition was founded approximately 50 years ago, during the early stages of computer science in this field. In the past few years, this field has made significant progress.
Author Information
Co-author: Dr. BAI Junxuan is a lecturer at IAIS of Capital University of Physical Education and Sports (CUPES). His main research interests are computer graphics and virtual reality. In 2021, he obtained his doctor’s degree from the State Key Laboratory of Virtual Reality Technology and Systems, School of Computer Science, Beihang University, and is now a team member of the Sports Biomechanics Research Center of IAIS, CUPES. He has published 14 papers in international mainstream journals and conferences in the field of computer science, including IEEE VR, IEEE ISMAR, ACM VRST, PG, DCC, CGF, TVC, CAVW, SCIENCE CHINA Information Sciences, and was authorized 4 patents. In January 2023, he was selected as a candidate for the 2023-2025 Youth Talent Promotion Project by Beijing Association for Science and Technology.
