Wei-Chiu Ma is a Ph.D. candidate at MIT, working with Antonio Torralba and Raquel Urtasun. His research lies in the intersection of computer vision, robotics, and machine learning, with a focus on in-the-wild 3D modeling and simulation and their applications to self-driving vehicles. Wei-Chiu is a recipient of the Siebel Scholarship and his work has been covered by media outlets such as WIRED, DeepLearning.AI, MIT News, etc. Previously, Wei-Chiu was a Sr. Research Scientist at Uber ATG R&D. He received his M.S. in Robotics from CMU where he was advised by Kris Kitani and B.S. in EE from National Taiwan University.
My research goal is to develop robust computational systems that can effectively perceive, model, and simulate the 3D world from unconstrained sensory data. To achieve this, my research program investigates the full spectrum of dynamic 3D world understanding: from robot localization to recognition, from static 3D reconstruction to dynamic motion estimation, and from closed-loop simulation to real-world robot manipulation. I examine these tasks not only in controlled settings, but also in sparse, noisy, and sometimes even extreme real-world settings where the systems will be deployed in. Specifically, I have studied how to combine the flexibility of deep neural networks with structured inductive biases, which allows us to significantly improve the performance, robustness, computational efficiency, and data dependency across a variety of 3D tasks and generalize to in-the-wild variation with minimal amounts of information. I have also explored how to construct composable, editable, and actionable 3D representations that allow robotic systems (e.g., self-driving vehicles, robot arms) to simulate counterfactual scenarios for better decision-making and act accordingly. Working with leading self-driving companies, I deploy my methods to improve the robustness and safety of existing autonomous systems.