4 papers
GeoDiff3D: Self-Supervised 3D Scene Generation with Geometry-Constrained 2D Diffusion Guidance
Haozhi Zhu, Miaomiao Zhao, Dingyao Liu +4
3D scene generation is a core technology for gaming, film/VFX, and VR/AR. Growing demand for rapid iteration, high-fidelity detail, and accessible content creation has further incr…
LL-Gaussian: Low-Light Scene Reconstruction and Enhancement via Gaussian Splatting for Novel View Synthesis
Hao Sun, Fenggen Yu, Huiyao Xu +2
Novel view synthesis (NVS) in low-light scenes remains a significant challenge due to degraded inputs characterized by severe noise, low dynamic range (LDR) and unreliable initiali…
CAGE-GS: High-fidelity Cage Based 3D Gaussian Splatting Deformation
Yifei Tong, Runze Tian, Xiao Han +3
As 3D Gaussian Splatting (3DGS) gains popularity as a 3D representation of real scenes, enabling user-friendly deformation to create novel scenes while preserving fine details from…
ARAP-GS: Drag-driven As-Rigid-As-Possible 3D Gaussian Splatting Editing with Diffusion Prior
Xiao Han, Runze Tian, Yifei Tong +3
Drag-driven editing has become popular among designers for its ability to modify complex geometric structures through simple and intuitive manipulation, allowing users to adjust an…