(SIGGRAPH 2016) Terrain-Adaptive Locomotion Skills Using Deep Reinforcement LearningMichiel van de Panne2016-05-03 | SIGGRAPH 2016 video for the related paper. The full paper can be found at http://www.cs.ubc.ca/~van/papers/2016-TOG-deepRL/index.htmlLearning to Steer (ICRA 2005)Michiel van de Panne2021-09-28 | "Learning to Steer on Winding Tracks Using Semi-Parametric Control Policies.", ICRA 2005: International Conference on Robotics and Automation Ken Alton, Michiel van de Panne. University of British Columbia. cs.ubc.ca/~van/papers/2005-icra-steering.pdfLearning to Steer and Back-upMichiel van de Panne2021-09-28 | Ken Alton and Michiel van de Panne "Learning to Steer on Winding Tracks Using Semi-Parametric Control Policies.", ICRA 2005, cs.ubc.ca/~van/papers/2005-icra-steering.pdfFlexible Motion Optimization with Modulated Assistive ForcesMichiel van de Panne2021-09-06 | Video supplementary material for: "Flexible Motion Optimization with Modulated Assistive Forces" Nam Hee Kim, Hung Yu Ling, Zhaoming Xie and Michiel van de Panne Symposium on Computer Animation, 2021 cs.ubc.ca/~van/papers/2021-SCA-modassist/index.htmlInteractive Control of Physically-based Animation: Luxo BasicsMichiel van de Panne2020-11-14 | ...Interactive Control for Physically-based Animation (SIGGRAPH 2000 paper)Michiel van de Panne2020-11-14 | Video that demonstrates some of the results the SIGGRAPH 2000 paper: J. F. Laszlo, M. van de Panne, and E. Fiume. Interactive Control For Physically-Based Animation. Proceedings of ACM SIGGRAPH 2000, p. 201-208.Learning to Locomote: Understanding How Environment Design Matters for Deep Reinforcement LearningMichiel van de Panne2020-10-17 | Video to accompany MIG 2020 paperSCA 2020: ALLSTEPS Curriculum-driven Learning of Stepping Stone Skills (full talk)Michiel van de Panne2020-09-28 | Full talk for the paper "ALLSTEPS: Curriculum-driven Learning of Stepping Stone Skills", ACM/EG Symposium on Computer Animation (SCA 2020). Project page: cs.ubc.ca/~van/papers/2020-allsteps/index.html Special thanks to STORROR Pro Parkour Team (storror.com) and Jesse Chui (songsforcinema.com)SCA 2020: ALLSTEPS Curriculum-driven Learning of Stepping Stone Skills (3 minute overview)Michiel van de Panne2020-09-28 | A short overview of the paper "ALLSTEPS: Curriculum-driven Learning of Stepping Stone Skills", ACM/EG Symposium on Computer Animation (SCA 2020). Project page: cs.ubc.ca/~van/papers/2020-allsteps/index.html Special thanks to STORROR Pro Parkour Team (storror.com), Jesse Chui (songsforcinema.com), and the Proclaimers (excerpts from "I'm Gonna Be (500 miles) the.proclaimers.co.uk/).[SIGGRAPH 2020] Character Controllers using Motion VAEsMichiel van de Panne2020-06-26 | Video for the related SIGGRAPH 2020 paper
A fundamental problem in computer animation is that of realizing purposeful and realistic human movement given a sufficiently-rich set of motion capture clips. We learn data-driven generative models of human movement using autoregressive conditional variational autoencoders, or Motion VAEs. The latent variables of the learned autoencoder define the action space for the movement and thereby govern its evolution over time. Planning or control algorithms can then use this action space to generate desired motions. In particular, we use deep reinforcement learning to learn controllers that achieve goal-directed movements. We demonstrate the effectiveness of the approach on multiple tasks. We further evaluate system-design choices and describe the current limitations of Motion VAEs.ALLSTEPS: Curriculum-driven Learning of Stepping Stone skillsMichiel van de Panne2020-05-11 | This video accompanies the paper of the same title. Curriculum-driven deep reinforcement learning is used to learn control policies for physics-based characters, enabling them to walk across stepping stone sequences and variable terrain. The control policy perceives the upcoming two desired stepping locations.Synthesis of Constrained Walking Skills (2008 SIGGRAPH ASIA)Michiel van de Panne2019-10-21 | Video supplementary materials for the 2008 SIGGRAPH ASIA paper cs.ubc.ca/~van/papers/2008-TOG-constrainedWalking/index.htmlauto completeMichiel van de Panne2018-09-26 | ...Feedback Control for Cassie with Deep Reinforcement LearningMichiel van de Panne2018-03-19 | video to accompany paper: arxiv.org/abs/1803.05580Progressive Reinforcement Learning with Distillation for Multi-Skilled Motion ControlMichiel van de Panne2018-03-04 | video accompanying ICLR 2018 paper: cs.ubc.ca/~van/papers/2018-ICLR-distill/2018-ICLR-distill.pdfMotion Doodles (SIGGRAPH 2004)Michiel van de Panne2018-03-04 | "Motion Doodles: An Interface for Sketching Character Motion" SIGGRAPH 2004. cs.ubc.ca/~van/papers/2004-siggraph-motiondoodles.pdf2005 sbim sketch3dMichiel van de Panne2018-03-04 | video accompanything paper: "Sketch-based Modeling of Parameterized Objects", 2nd Eurographics Workshop on Sketch-Based Interfaces and Modeling, 2005. cs.ubc.ca/~van/papers/2005-sbim-sketch3d.pdfSIGGRAPH 2017 DeepLoco (supplemental video)Michiel van de Panne2017-04-28 | Supplemental video accompanying the SIGGRAPH 2017 paper: "DeepLoco: Dynamic Locomotion Skills Using Hierarchical Deep Reinforcement Learning". Project page: http://www.cs.ubc.ca/~van/papers/2017-TOG-deepLoco/index.htmlSIGGRAPH 2017: DeepLoco paper (main video)Michiel van de Panne2017-04-28 | Main video accompanying the SIGGRAPH 2017 paper: "DeepLoco: Dynamic Locomotion Skills Using Hierarchical Deep Reinforcement Learning". Project page: http://www.cs.ubc.ca/~van/papers/2017-TOG-deepLoco/index.html[SIGGRAPH 2017] DeepLoco: HighlightsMichiel van de Panne2017-04-28 | Highlights video for the SIGGRAPH 2017 paper. Project page: http://www.cs.ubc.ca/~van/papers/2017-TOG-deepLoco/index.html