Changchang Liu

Beijing University of Civil Engineering and Architecture

Papers

Total Citations

H-Index

About

Changchang Liu is a leading researcher in the intersection of computer vision, robotics, and semantic mapping, with a primary focus on advancing Visual Simultaneous Localization and Mapping (vSLAM) in complex, dynamic environments. Her major contributions center on integrating semantic information into traditional SLAM frameworks to overcome the limitations of static-scene assumptions. In her highly cited 2023 work, "A Semantic Information-Based Optimized vSLAM in Indoor Dynamic Environments" (10 citations), Liu pioneered methods for enabling mobile robots to accurately localize and build dense maps even amidst moving objects, a critical step toward real-world robotic autonomy. She further advanced this field with her 2025 paper on "DeepLabV3+-Based Semantic Annotation Refinement for SLAM," which addresses the challenge of reconstructing 3D scenes from monocular imagery in semantically sparse settings, significantly boosting robotic operational efficiency. By fusing deep learning segmentation with geometric mapping, Liu’s research provides robust solutions for indoor navigation, directly impacting the development of intelligent service robots and autonomous systems. Her work stands out for its practical approach to bridging the gap between theoretical SLAM and real-world deployment.