Xiaoxiang Cao
Papers
1
Total Citations
9
H-Index
1
About
Xiaoxiang Cao is a researcher at the forefront of visual simultaneous localization and mapping (SLAM), with a primary focus on enhancing robotic perception in dynamic, real-world environments. His key research areas span computer vision, deep learning, and autonomous navigation, where he addresses the critical challenge of robust feature management in changing scenes. Cao’s most notable contribution is the development of GAT-LSTM, a pioneering network that integrates graph attention mechanisms with long short-term memory to intelligently select and manage feature points for visual SLAM. This work, published in 2025 and already garnering 9 citations, offers a significant leap over traditional static-environment assumptions, enabling robots to maintain accurate localization even amidst moving objects. By fusing graph-based relational reasoning with temporal sequence learning, his approach improves both the stability and efficiency of feature tracking, directly impacting applications in autonomous driving, service robotics, and augmented reality. Cao’s research stands out for its elegant synthesis of geometric and learning-based methods, providing a practical pathway toward more resilient and adaptive robotic systems. His ongoing work promises to further bridge the gap between controlled laboratory settings and the unpredictable dynamics of the physical world.
Research Focus
Key Achievements
Top Papers
- 1