Research map / 3D vision / Gaussian splatting
TrackGS: Optimizing COLMAP-Free 3D Gaussian Splatting with Global Track Constraints
Dongbo Shi, Shen Cao, Lubin Fan, Bojian Wu, Jinhui Guo, Ligang Liu, Renjie Chen
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 2D track loss (ablation) | TrackGS · Synthetic (4 scenes, avg) | PSNR: 18.18 -> 32.58 | Improved |
| add 2D track loss (ablation) | TrackGS · Synthetic (4 scenes, avg) | FoV error: 2.617 -> 0.015 | Improved |
| add 3D track loss (ablation) | TrackGS · Synthetic (4 scenes, avg) | PSNR: 32.38 -> 32.58 | Improved |
| add 3D track loss (ablation) | TrackGS · Synthetic (4 scenes, avg) | ATE: 0.00275 -> 0.00018 | Improved |
| optimize camera intrinsics instead of fixed 60 degree FoV (ablation) | TrackGS · Synthetic (4 scenes) | PSNR: 24.01 -> 31.17 (bedroom) | Improved |
| use TrackGS instead of COLMAP-assisted 3DGS | TrackGS · Synthetic (4 scenes) | PSNR: 31.73 -> 31.17 (bedroom), 35.81 -> 36.26 (classroom) | Mixed |
| use global track constraints instead of local constraints (CF-3DGS) | TrackGS · CO3D-V2 (scene 46_2587_7531) | ATE: 0.009 -> 0.001 | Improved |
| use TrackGS vs CF-3DGS on Tanks and Temples Ignatius | TrackGS · Tanks and Temples Ignatius | PSNR: 14.96 -> 22.12 | Improved |
Closest papers by meaning
- GaussianSpa: An "Optimizing-Sparsifying" Simplification Framework for Compact and High-Quality 3D Gaussian Splatting (CVPR 2025 · similarity 0.92)
- Steepest Descent Density Control for Compact 3D Gaussian Splatting (CVPR 2025 · similarity 0.91)
- GaussianPro: 3D Gaussian Splatting with Progressive Propagation (ICML 2024 · similarity 0.91)
- COLMAP-Free 3D Gaussian Splatting (CVPR 2024 · similarity 0.91)
- Speedy-Splat: Fast 3D Gaussian Splatting with Sparse Pixels and Sparse Primitives (CVPR 2025 · similarity 0.91)
- SplatFormer: Point Transformer for Robust 3D Gaussian Splatting (ICLR 2025 · similarity 0.91)
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.
