Research map / 3D vision / Gaussian splatting
LongSplat: Robust Unposed 3D Gaussian Splatting for Casual Long Videos
Chin-Yang Lin, Cheng Sun, Fu-En Yang, Min-Hung Chen, Yen-Yu Lin, Yu-Lun Liu
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 camera pose estimation (ablation) | LongSplat · Free dataset (avg) | PSNR: 20.19 -> 27.88 | Improved |
| add global optimization (ablation) | LongSplat · Free dataset (avg) | PSNR: 20.50 -> 27.88 | Improved |
| add local optimization (ablation) | LongSplat · Free dataset (avg) | PSNR: 25.94 -> 27.88 | Improved |
| add final refinement stage (ablation) | LongSplat · Free dataset (avg) | PSNR: 26.08 -> 27.88 | Improved |
| use visibility-adaptive window instead of 5-frame fixed window (ablation) | LongSplat · Free dataset (avg) | PSNR: 26.90 -> 27.88 | Improved |
| use visibility-adaptive window instead of optimizing all frames globally (ablation) | LongSplat · Free dataset (avg) | PSNR: 26.15 -> 27.88 | Improved |
| use adaptive octree anchors instead of per-pixel unprojection (ablation) | LongSplat · Free dataset (avg) | PSNR: 22.47 -> 27.88 | Improved |
| use adaptive octree anchors instead of fixed-size voxels (ablation) | LongSplat · Free dataset (avg) | PSNR: 26.99 -> 27.88 | Improved |
| use adaptive octree anchors instead of naive densification (ablation) | LongSplat · Free dataset (avg) | PSNR: 25.73 -> 27.88 | Improved |
| add 2D correspondence loss (ablation) | LongSplat · Free dataset (avg) | PSNR: 26.54 -> 27.88 | Improved |
| add depth loss (ablation) | LongSplat · Free dataset (avg) | PSNR: 26.74 -> 27.88 | Improved |
| jointly optimize MASt3R-initialized poses | MASt3R + Scaffold-GS · Free dataset (avg) | PSNR: 23.05 -> 24.42 | Improved |
Later papers that compared with this method
2 later accepted papers report LongSplat as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.
- ARTDECO: Toward High-Fidelity On-the-Fly Reconstruction with Hierarchical Gaussian Structure and Feed-Forward Guidance (ICLR 2026 · better on 7 of 7 shared results)
- AeroGS: Scale-Aware Gaussian Splatting for Pose-Free Dynamic UAV Scene Reconstruction (CVPR 2026 · better on 3 of 3 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- NoPe-NeRF — NoPe-NeRF: Optimising Neural Radiance Field With No Pose Prior (CVPR 2023) · 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
- SelfSplat: Pose-Free and 3D Prior-Free Generalizable 3D Gaussian Splatting (CVPR 2025 · similarity 0.91)
- No Pose, No Problem: Surprisingly Simple 3D Gaussian Splats from Sparse Unposed Images (ICLR 2025 · similarity 0.91)
- FreeSplat: Generalizable 3D Gaussian Splatting Towards Free View Synthesis of Indoor Scenes (NeurIPS 2024 · similarity 0.90)
- PF3plat: Pose-Free Feed-Forward 3D Gaussian Splatting for Novel View Synthesis (ICML 2025 · similarity 0.90)
- KeyGS: A Keyframe-Centric Gaussian Splatting Method for Monocular Image Sequences (AAAI 2025 · similarity 0.90)
- MonoSplat: Generalizable 3D Gaussian Splatting from Monocular Depth Foundation Models (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.
