Max Robotics — Global Service Robot Directory + US Certification

About

Xinghui Dong is a computer vision researcher whose work spans autonomous inspection, facial analysis, and material perception. His most cited paper, "Ship hull inspection: A survey" (2023, 49 citations), provides a comprehensive overview of robotic and vision-based methods for underwater hull assessment, establishing him as a key voice in maritime inspection. In photometric stereo, Dong introduced a dual-cue network (2019, 27 citations) that fuses multispectral cues to recover surface normals, advancing 3D reconstruction for challenging materials. His contributions to affective computing include a cascade regression-based face frontalization method (2021, 23 citations) that improves dynamic facial expression recognition by synthesizing frontal views from non-frontal inputs—critical for human-computer interaction and healthcare applications. Dong also tackled the difficult problem of transparent object detection, leveraging instance segmentation (2019, 11 citations) to identify glass and plastic in cluttered scenes, with implications for robotics and autonomous navigation. Earlier work on monocular visual-inertial odometry (2016, 6 citations) evaluated detector-descriptor pipelines for state estimation. Across these projects, Dong demonstrates a talent for addressing real-world perception challenges—from underwater hulls to transparent materials—with practical, learning-based solutions that have earned steady citation growth.