Korrawe Karunratanakul
Profile photo

Hi! I am Korrawe Karunratanakul (กรวีร์ การุณรัตนกุล). Currently, I am a postdoctoral researcher at ETH Zurich working with Prof. Siyu Tang, who was also my PhD advisor. I was co-advised by Prof. Otmar Hilliges in the early years of my PhD. Before that, I was a research intern at MPI-IS supervised by Siyu and Krikamol Muandet in 2019. I obtained my master's and bachelor's degrees from Chulalongkorn University, Thailand, advised by Dr. Ekapol Chuangsuwanich and Dr. Sira Sriswasdi.

I aspire to advance our understanding of how humans interact with the 3D world and leverage it to improve our daily life. My current research interests lie in the intersection of computer vision, machine learning, and robotics, with a focus on generative models for motion modeling and robotics, world models, and efficient learning methods from 2D/3D data.

I am open to collaboration opportunities that align well with my interests and background. I am also looking for an interesting role in the industry after my postdoc. Please feel free to reach out!

News
Publications
UniPhys: Unified Planner and Controller with Diffusion for Flexible Physics-Based Character Control
ControlMM: Controllable Masked Motion Generation
DNO: Optimizing Diffusion Noise Can Serve As Universal Motion Priors
GMD: Controllable Human Motion Synthesis via Guided Diffusion Models
HARP: Personalized Hand Reconstruction from a Monocular RGB Video
Conference on Computer Vision and Pattern Recognition (CVPR), 2023
Korrawe Karunratanakul, Sergey Prokudin, Otmar Hilliges, Siyu Tang
HALO: A Skeleton-Driven Neural Occupancy Representation for Articulated Hands
Reducing Spelling Inconsistencies in Code-Switching ASR using Contextualized CTC Loss
IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2021
Grasping Field: Learning Implicit Representations for Human Grasps
International Conference on 3D Vision (3DV), 2020
(Best Paper Award)
Uncovering thousands of new peptides with sequence-mask-search hybrid de novo peptide sequencing framework