Jingwei Yang

Papers

Total Citations

H-Index

About

Jingwei Yang is a researcher in computer vision and graphics, with a particular focus on depth-to-normal translation and surface normal estimation. Their most notable contribution is the "Three-Filters-to-Normal+" (3F2N+) framework, introduced in 2024, which revisits and improves discontinuity discrimination in depth-to-normal conversion. This work addresses a fundamental challenge in 3D scene understanding: accurately inferring surface normals from depth maps, especially at object boundaries and sharp edges where traditional methods often fail. By refining the filtering process to better preserve geometric discontinuities, Yang's approach enhances the quality of normal maps used in applications like 3D reconstruction, rendering, and robotic perception. Though early in its citation trajectory with 3 citations, the work represents a meaningful step forward in a core computer vision task. Yang's research sits at the intersection of geometric deep learning and classical image processing, aiming to bridge the gap between raw sensor data and high-quality 3D representations. Their work is particularly relevant for students and researchers interested in differentiable rendering, neural scene representations, and robust geometric feature extraction from noisy depth inputs.