Jinyan Ni
Papers
1
Total Citations
7
H-Index
1
About
Jinyan Ni is a researcher whose work centers on advancing computer vision, particularly in the domain of stereo matching for three-dimensional reconstruction and robotic perception. Their most-cited paper, "Stereo matching using census cost over cross window and segmentation-based disparity refinement" (2018, 7 citations), introduces a practical method that enhances depth estimation accuracy by combining a census cost function with adaptive cross-window support and segmentation-based refinement. This contribution addresses critical challenges in applications like robot navigation, object detection, and industrial measurement, where reliable 3-D information is essential. By improving the robustness of disparity maps in complex scenes, Ni’s work helps bridge the gap between theoretical stereo algorithms and real-world deployment. Though their citation impact is still growing, the focus on segmentation-driven refinement marks a notable step toward more efficient and accurate depth sensing. For students and researchers exploring stereo vision, Ni’s approach offers a clear example of how integrating cost aggregation with post-processing can yield tangible performance gains in practical systems.
Research Focus
Key Achievements
Top Papers
- 1