Jae-Seong Yun
Papers
1
Total Citations
2
H-Index
1
About
Jae-Seong Yun is a computer vision researcher whose work centers on novel-view synthesis and 3D scene representation, with a particular focus on enabling robust rendering from ground-robot perspectives. His key contribution, the Mode-GS algorithm, tackles a critical limitation in existing 3D Gaussian splatting methods: the severe splat drift that occurs when rendering scenes from low, ground-level camera trajectories. By introducing monocular depth-guided anchored Gaussian splats, Yun’s approach stabilizes the 3D representation, dramatically improving rendering quality for autonomous navigation and robotics applications. Though his most-cited paper, "Mode-GS: Monocular Depth Guided Anchored 3D Gaussian Splatting for Robust Ground-View Scene Rendering" (2024), has garnered 2 citations in its early stage, the work represents a timely and practical advance in neural rendering—addressing a real-world deployment challenge that prior methods overlooked. Yun’s research sits at the intersection of 3D computer vision, neural rendering, and robotics perception, and his anchored splat methodology offers a promising direction for future work in view synthesis under constrained camera poses.
Research Focus
Key Achievements
Top Papers
- 1