Software

  • RoboMeshCat, a set of utilities for visualizing robots in web-based visualizer MeshCat is available on [GitHub], [Conda] and [PyPI].
  • PyPhysX, python wrapper for Nvidia PhysX simulator is on [GitHub].

Recent contributions

  • August, 2022: Learning Object Manipulation Skills from Video via Approximate Differentiable Physics. Vladimir Petrik, Mohammad Nomaan Qureshi, Josef Sivic, and Makarand Tapaswi. Paper was accepted for publication at IROS conference and is available on [ProjectPage], [YouTube], [GitHub], or [arXiv].

  • July, 2022: Collision Detection Accelerated: An Optimization Perspective. Louis Montaut, Quentin Le Lidec, Vladimir Petrik, Josef Sivic, Justin Carpentier. Paper was presented at RSS conference and is available on [RSS] or [arXiv].

  • January, 2022: Learning to Manipulate Tools by Aligning Simulation to Video Demonstration. Kateryna Zorina, Justin Carpentier, Josef Sivic, Vladimír Petrík. Paper was published in IEEE Robotics and Automation Letters journal [IEEE] and the work received the best poster award [PosterAward] at 2nd International Workshop on AI for Robotics. The work is available at [ProjectPage] or [arXiv].

  • November, 2020: Learning Object Manipulation Skills via Approximate State Estimation from Real Videos. Vladimír Petrík, Makarand Tapaswi, Ivan Laptev, Josef Šivic. Paper was accepted and presented at CoRL conference and is available on [arXiv], [youtube], or [GitHub].