Cong Yang
Papers
1
Total Citations
3
H-Index
1
About
Cong Yang is a computer vision researcher whose work centers on geometric understanding for robotics and autonomous systems. His most notable contribution is the development of SiTPose, a Siamese convolutional transformer that directly regresses relative camera pose from image pairs—a fundamental challenge in visual odometry and 3D reconstruction. Unlike traditional pipelines that rely on feature matching or iterative refinement, Yang’s model achieves end-to-end pose estimation by fusing convolutional and transformer architectures, offering a more efficient and robust solution for real-time applications. Though his 2023 paper has garnered 3 citations in its early stage, the work demonstrates a promising direction for bridging deep learning with geometric computer vision. Yang’s research addresses the critical need for accurate spatial reasoning in robotics, particularly in scenarios with limited overlap or dynamic environments. By simplifying the relative camera pose estimation pipeline, his approach has the potential to impact fields ranging from augmented reality to drone navigation, where rapid and reliable pose inference is essential.
Research Focus
Key Achievements
Top Papers
- 1