Jingwei Ji
Papers
1
Total Citations
5
H-Index
1
About
Jingwei Ji is a researcher whose work lies at the intersection of computer vision and 3D geometric deep learning, with a particular focus on reconstructing three-dimensional objects from limited visual data. His most cited contribution, "DeformNet: Free-Form Deformation Network for 3D Shape Reconstruction from a Single Image" (2018), addresses a fundamental challenge in fields ranging from robotic manipulation to augmented reality. While prior approaches relied on generative models that output voxel grids or point clouds—often at high computational cost—Ji introduced a novel paradigm using free-form deformation. This technique allows the network to deform a pre-defined 3D template mesh to match the object in a single 2D image, achieving efficient and high-quality shape reconstruction. Though his citation count is still building, this work has been recognized for its elegant solution to a persistent problem in 3D vision. Ji’s research is notable for its practical implications: enabling robots to better understand their environment and powering more immersive AR experiences. His approach represents a significant step toward making 3D reconstruction from minimal input both accurate and computationally feasible.
Research Focus
Key Achievements
Top Papers
- 1