Shengke Wang

Papers

Total Citations

143

H-Index

About

Shengke Wang is a computer vision researcher whose work centers on deep learning-based object detection, with a particular specialization in the challenging domain of underwater image analysis. His research addresses a critical gap in the field: while conventional deep learning detection methods perform admirably in controlled environments, they struggle significantly when confronted with the unique degradation factors present in underwater imagery, including poor visibility, small object sizes, and high levels of noise and blur. Wang's most influential contribution, "Underwater Object Detection using Invert Multi-Class Adaboost with Deep Learning" (2020), has garnered 118 citations, establishing him as a notable voice in this specialized niche. By integrating Invert Multi-Class Adaboost techniques with modern deep learning architectures, he developed methods capable of more robustly handling the complexities that standard detectors fail to address. His subsequent work, SWIPENET, further refined these capabilities by specifically targeting noise robustness in underwater scenes. Wang's research carries meaningful real-world implications for marine biology, autonomous underwater vehicles, and ocean exploration technologies. His collective body of work signals a dedicated focus on pushing the boundaries of object detection into some of computer vision's most demanding environmental conditions.