Dawei Yang

Papers

Total Citations

H-Index

About

Dawei Yang is a computer vision researcher whose work centers on object detection, image segmentation, and visual perception systems for robotics applications. His research addresses fundamental challenges in enabling machines to accurately interpret complex visual environments, with particular emphasis on practical deployment scenarios involving mobile robots. Yang's most recognized contribution, "Convolutional Feature Frequency Adaptive Fusion Object Detection Network" (2021), has accumulated 9 citations and advances multi-scale feature fusion strategies in convolutional neural network architectures, improving detection accuracy across varying object scales. His earlier work on cascaded superpixel pedestrian segmentation (2018) tackled a persistent challenge in mobile robotics — accurately delineating human body contours in cluttered indoor and outdoor environments where background interference degrades performance. By incorporating superpixel-based cascaded processing to account for background noise, Yang developed a more robust segmentation pipeline suited for real-world robotic vision systems. Yang's research reflects a consistent focus on bridging theoretical computer vision advances with practical robotic applications, particularly in safety-critical scenarios involving human detection and scene understanding. While his citation profile is still developing, his methodological contributions to adaptive feature fusion and semantically-aware segmentation represent meaningful progress in robust visual perception systems.