Research map / 3D vision / Gaussian splatting
PCR-GS: COLMAP-Free 3D Gaussian Splatting via Pose Co-Regularizations
Yu Wei, Jiahui Zhang, Xiaoqin Zhang, Ling Shao, Shijian Lu
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 feature reprojection regularization (DINO) (ablation) | CF-3DGS · Tanks&Temples ablation (4 fps) | PSNR: 18.34 -> 23.16 | Improved |
| add feature reprojection regularization (DINO) (ablation) | CF-3DGS · Tanks&Temples ablation (4 fps) | LPIPS: 0.32 -> 0.17 | Improved |
| add wavelet regularization without high-frequency components (ablation) | CF-3DGS · Tanks&Temples ablation (4 fps) | PSNR: 18.34 -> 18.40 | No clear effect |
| add wavelet regularization with high-frequency components (ablation) | CF-3DGS · Tanks&Temples ablation (4 fps) | PSNR: 18.34 -> 19.31 | Improved |
| include high-frequency components in wavelet regularization (ablation) | CF-3DGS+WRF · Tanks&Temples ablation (4 fps) | PSNR: 18.40 -> 19.31 | Improved |
| add wavelet-based frequency regularization on top of feature reprojection (ablation) | CF-3DGS+FRR · Tanks&Temples ablation (4 fps) | PSNR: 23.16 -> 24.20 | Improved |
| add wavelet-based frequency regularization on top of feature reprojection (ablation) | CF-3DGS+FRR · Tanks&Temples ablation (4 fps) | SSIM: 0.72 -> 0.79 | Improved |
| replace CF-3DGS with PCR-GS | CF-3DGS · Tanks&Temples (8 scenes, 4 fps) mean | PSNR: 19.79 -> 23.68 | Improved |
| replace CF-3DGS with PCR-GS | CF-3DGS · Tanks&Temples (8 scenes, 4 fps) mean | RPE_t: 0.211 -> 0.109 | Improved |
| replace CF-3DGS with PCR-GS | CF-3DGS · Tanks&Temples (8 scenes, 4 fps) mean | RPE_r: 0.520 -> 0.350 | Improved |
| replace CF-3DGS with PCR-GS | CF-3DGS · Free-Dataset (3 scenes) mean | PSNR: 15.00 -> 17.78 | Improved |
| replace CF-3DGS with PCR-GS | CF-3DGS · Free-Dataset (3 scenes) mean | ATE: 0.042 -> 0.024 | Improved |
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.
- Nope-NeRF — NoPe-NeRF: Optimising Neural Radiance Field With No Pose Prior (CVPR 2023) · better on 6 of 6 shared results
- BARF — BARF: Bundle-Adjusting Neural Radiance Fields (ICCV 2021) · better on 2 of 2 shared results
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
- KeyGS: A Keyframe-Centric Gaussian Splatting Method for Monocular Image Sequences (AAAI 2025 · similarity 0.88)
- SfM-Free 3D Gaussian Splatting via Hierarchical Training (CVPR 2025 · similarity 0.88)
- SDGS: Spatial Difference Guided Gaussian Splatting for Simultaneous Localization and 3D Reconstruction (CVPR 2026 · similarity 0.88)
- SelfSplat: Pose-Free and 3D Prior-Free Generalizable 3D Gaussian Splatting (CVPR 2025 · similarity 0.86)
- COLMAP-Free 3D Gaussian Splatting (CVPR 2024 · similarity 0.86)
- GS-CPR: Efficient Camera Pose Refinement via 3D Gaussian Splatting (ICLR 2025 · similarity 0.86)
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.
