Research map / 3D vision / Gaussian splatting
RDG-GS: Relative Depth Guidance with Gaussian Splatting for Real-time Sparse-View 3D Rendering
Chenlu Zhan, Yufei Zhang, Yu Lin, Gaoang Wang, Hongwei Wang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add refined depth to the 3D-GS baseline (ablation) | 3D-GS · Mip-NeRF360 (1/8 res) | PSNR: 20.89 -> 22.68 | Improved |
| use refined depth instead of coarse DPT depth (ablation) | 3D-GS · Mip-NeRF360 (1/8 res) | PSNR: 21.43 -> 22.68 | Improved |
| add relative depth guidance on top of refined depth (ablation) | RDG-GS · Mip-NeRF360 (1/8 res) | PSNR: 22.68 -> 24.81 | Improved |
| add relative depth guidance on top of refined depth (ablation) | RDG-GS · NeRF-LLFF 1006x762 | PSNR: 18.67 -> 20.22 | Improved |
| add adaptive sampling on top of refined depth (ablation) | RDG-GS · Mip-NeRF360 (1/8 res) | PSNR: 22.68 -> 25.11 | Improved |
| use refined depth with RDG and AS instead of coarse depth (ablation) | RDG-GS · Mip-NeRF360 (1/8 res) | PSNR: 24.38 -> 26.03 | Improved |
| use adaptive sampling instead of random sampling initialization (ablation) | RDG-GS · Mip-NeRF360 (1/4 res) | PSNR: 19.15 -> 25.01 | Improved |
| use adaptive sampling instead of COLMAP dense initialization (ablation) | RDG-GS · Mip-NeRF360 (1/4 res) | PSNR: 21.76 -> 25.01 | Improved |
| use adaptive sampling instead of Gaussian Unpooling initialization (ablation) | RDG-GS · DTU | PSNR: 21.21 -> 23.51 | Improved |
| switch DPT depth model from hybrid to large (ablation) | FSGS · Mip-NeRF360 (1/2 res) | PSNR: 19.56 -> 20.11 | Mixed |
| switch DPT depth model from hybrid to large (ablation) | RDG-GS · Mip-NeRF360 (1/2 res) | PSNR: 22.62 -> 22.67 | No clear effect |
| raise SH degree from 2 to 3 (ablation) | RDG-GS · NeRF-LLFF 503x381 | PSNR: 22.01 -> 22.03 | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- CoR-GS — CoR-GS: Sparse-View 3D Gaussian Splatting via Co-Regularization (ECCV 2024) · better on 4 of 4 shared results
- SparseNeRF — SparseNeRF: Distilling Depth Ranking for Few-shot Novel View Synthesis (ICCV 2023) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- D$^2$GS: Depth-and-Density Guided Gaussian Splatting for Stable and Accurate Sparse-View Reconstruction (ICLR 2026 · similarity 0.92)
- NexusGS: Sparse View Synthesis with Epipolar Depth Priors in 3D Gaussian Splatting (CVPR 2025 · similarity 0.91)
- 3D Convex Splatting: Radiance Field Rendering with 3D Smooth Convexes (CVPR 2025 · similarity 0.90)
- PUP 3D-GS: Principled Uncertainty Pruning for 3D Gaussian Splatting (CVPR 2025 · similarity 0.90)
- MS-GS: Multi-Appearance Sparse-View 3D Gaussian Splatting in the Wild (NeurIPS 2025 · similarity 0.90)
- Speedy-Splat: Fast 3D Gaussian Splatting with Sparse Pixels and Sparse Primitives (CVPR 2025 · similarity 0.90)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
