Max Robotics — Global Service Robot Directory + US Certification

About

Cheng-Hao Kuo’s research lies at the intersection of computer vision, robotics, and affective computing, with a focus on enabling machines to perceive and understand the world in 3D. He has made significant contributions to monocular depth estimation, body orientation estimation, and neural scene representations. His work on MEBOW (Monocular Estimation of Body Orientation in the Wild) introduced a novel approach for estimating body orientation from single images, a critical capability for autonomous driving and robotics, and has garnered 40 citations. Kuo has also advanced neural radiance fields (NeRF) by developing a multimodal framework that integrates multiple sensory modalities for enhanced scene reconstruction, and he has tackled the challenge of domain adaptation in monocular depth estimation through feature decomposition. Beyond geometric perception, Kuo explores affective computing, proposing a semi-supervised metric learning framework for emotion recognition from video. His most recent work on continuous semantic splatting models uncertainty in 3D Gaussian splatting, pushing the boundaries of probabilistic scene understanding. With a growing citation record and a portfolio spanning from robust visual perception to socially intelligent systems, Kuo is a rising figure in applied computer vision.