Changsheng Lu

Australian National University, Shanghai Jiao Tong University

Papers

Total Citations

H-Index

About

Changsheng Lu is a researcher specializing in computer vision and robotics, with a primary focus on viewpoint estimation—a critical pre-procedure for purposive perception and fine pose estimation in robotic manipulation and grasping. His work addresses the fundamental challenge of enabling robots to determine optimal camera viewpoints relative to objects, which significantly enhances visual system performance in both observation and manipulation tasks. Lu’s major contributions include developing a domain adaptation approach for viewpoint estimation using image generation (2021, 6 citations), which tackles the persistent problem of limited training data with accurate annotations. He also pioneered a viewpoint estimation method employing triplet loss with a novel viewpoint-based input selection strategy (2019, 2 citations), leveraging CNN-based algorithms to extract discriminative features effectively. These innovations have direct applications in advancing robot manipulation and grasping capabilities. While his citation counts are still growing, Lu’s work represents foundational steps toward more robust and data-efficient viewpoint estimation systems, positioning him as an emerging contributor to the intersection of computer vision and robotics. His research continues to push boundaries in enabling more intelligent and perceptive robotic systems.