DOI QR코드

DOI QR Code

Research on Effective Feature Vector Configuration for Motion Matching in Locomotive Motion Generation

보행 동작 생성을 위한 모션 매칭의 효과적인 특징 벡터 설정에 관한 연구

  • Sura Kim (Department of Software, Sejong University) ;
  • Sang Il Park (Department of Software, Sejong University)
  • 김수라 (세종대학교 소프트웨어학과) ;
  • 박상일 (세종대학교 소프트웨어학과)
  • Received : 2023.06.18
  • Accepted : 2023.07.05
  • Published : 2023.07.25

Abstract

This paper investigates effective methods for implementing motion matching, which is actively used in real-time motion generation applications. The success of motion matching heavily hinges on its simple definition of a feature vector, yet this very definition can introduce significant variance in the outcomes. Our research focuses on identifying the optimal combination of feature vectors that effectively generates desired trajectories in locomotion generation. To this end, we experimented with a range of feature vector combinations and performed an in-depth error analysis to evaluate the results.

이 논문은 실시간 모션 생성 응용분야에서 최근 활발하게 사용되고 있는 모션 매칭 기술을 효과적으로 적용하는 방법에 대해 연구한다. 모션 매칭의 핵심은 간단한 특징 벡터의 정의에 있으나 이 정의의 선택에 따라 결과가 많이 달라지게 된다. 이에 본 연구에서는 보행 동작을 중심으로 사용자가 원하는 궤적을 생성하기 위한 최적의 특징 벡터의 조합을 도출하는 것을 목표로 다양한 조합에 대해 실험하고, 이에 대한 오차 분석을 시행하였다.

Keywords

References

  1. L. Kovar, M. Gleicher, and F. Pighin, Motion graphs, ACM Transactions on Graphics 21, 3, 473-482. 2002  https://doi.org/10.1145/566654.566605
  2. J. Lee, J. Chai, P. Reitsma, J. Hodgins and N. Pollard, Interactive control of avatars animated with human motion data. ACM Transactions on Graphics 21, 3, 491-500, 2002  https://doi.org/10.1145/566654.566607
  3. O. Arikan, and D. Forsythe, Interactive motion generation from examples. ACM Transactions on Graphics 21, 3, 483-490, 2002  https://doi.org/10.1145/566654.566606
  4. L. Kovar and M. Gleicher, Automated extraction and parameterization of motions in large data sets. ACM Trans. Graph. 23(3): 559-568, 2004  https://doi.org/10.1145/1015706.1015760
  5. K. Hyun, K. Lee, and J. Lee, Motion Grammars for Character Animation. In Proceedings of the 37th Annual Conference of the European Association for Computer Graphics (EG '16). 103-113, 2016 
  6. Y. Lee, K. Wampler, G. Bernstein, J. Popovic, and Z. Popovic, Motion Fields for Interactive Character Locomotion. In ACM SIGGRAPH Asia 2010, Article 138, 2010 
  7. K. Lee, S. Lee, and J. Lee, Interactive character animation by learning multi-objective control, ACM Trans. Graph., vol. 37, no. 6, pp. 1-10, Dec. 2018.  https://doi.org/10.1145/3272127.3275071
  8. K. Lee, S. Min, S. Lee, and J. Lee, Learning time-critical responses for interactive character control, ACM Trans. Graph., vol. 40, no. 4, pp. 1-11, Aug. 2021.  https://doi.org/10.1145/3476576.3476726
  9. K. Chen, Z. Tan, J. Lei, S.-H. Zhang, Y.-C. Guo, W. Zhang, and S.-M. Hu, ChoreoMaster: Choreography-oriented music-driven dance synthesis, ACM Trans. Graph., vol. 40, no. 4, pp. 1-13, Aug. 2021.  https://doi.org/10.1145/3476576.3476724
  10. K. Cho, C. Kim, J. Park, J. Park, and J. Noh, Motion recommendation for online character control, ACM Trans. Graph., vol. 40, no. 6, pp. 1-16, 2021  https://doi.org/10.1145/3478513.3480512
  11. M. Buttner and S. Clavet, Motion Matching - The Road to Next Gen Animation. In Proc. of Nucl.ai 2015, 2015 
  12. S. Clavet. Motion Matching and The Road to Next-Gen Animation. In Proc. of GDC 2016, 2016 
  13. K. Zadziuk. Motion Matching, The Future of Games Animation... Today. In Proc. of GDC 2016, 2016 
  14. G. Harrower. Real Player Motion Tech in 'EA Sports UFC 3'. In Proc. of GDC 2018, 2018 
  15. M. Buttner, Machine Learning for Motion Synthesis and Character Control in Games. In Proc. of i3D 2019, 2019 
  16. F. Zinno, ML Tutorial Day: From Motion Matching to Motion Synthesis, and All the Hurdles In Between. In Proc. of GDC 2019. 2019 
  17. D. Holden. 2018. Character Control with Neural Networks and Machine Learning. In Proc. of GDC 2018, 2018 
  18. D. Holden, O. Kanoun, M. Perepichka, and T. Popa, Learned motion matching, ACM Transactions on Graphics, 39, 2020 
  19. J. Lee, T. Kwon and Y. Lee, "Interactive Character Path-Following Using Long-Horizon Motion Matching With Revised Future Queries," in IEEE Access, vol. 11, pp. 9942-9956, 2023  https://doi.org/10.1109/ACCESS.2023.3240589
  20. J. Lee and S. Y. Shin, "A hierarchical approach to interactive motion editing for human-like figures." In Proceedings of SIGGRAPH '99, 39-48, 1999 
  21. https://theorangeduck.com/page/phase-functioned-neural-networks-character-control