Yusheng Wang

Papers

Total Citations

H-Index

About

Yusheng Wang is a rising researcher in underwater robotics and computer vision, whose work addresses the fundamental challenge of extracting 3D information from 2D forward-looking sonar (FLS) imagery. His key research areas include multi-view stereo reconstruction, acoustic image processing, and sonar simulation. Wang’s major contribution is his pioneering approach to retrieving missing depth information from 2D sonar images—a well-known problem in the field. In his most-cited work, "Learning Pseudo Front Depth for 2D Forward-Looking Sonar-based Multi-view Stereo" (2022, 21 citations), he introduces a learning-based method that enables robots to generate 3D maps during fly-through motion, effectively overcoming the inherent dimensionality loss in acoustic imaging. His subsequent work, "2D Forward Looking Sonar Simulation with Ground Echo Modeling" (2023, 5 citations), addresses the critical need for realistic simulation environments, incorporating ground echo effects to improve the fidelity of sonar data for training and testing algorithms. This simulation work is particularly valuable given the difficulty and expense of real-world underwater data collection. Wang’s research is directly applicable to autonomous underwater vehicles (AUVs) operating in turbid or low-light conditions where optical cameras fail, positioning him as an important contributor to advancing perception capabilities in underwater robotics.