Research map / 3D vision / Gaussian splatting
LongSplat: Online Generalizable 3D Gaussian Splatting from Long Sequence Images
Guichen Huang, Ruoyu Wang, Xiangjun Gao, Che Sun, Yuwei Wu, Shenghua Gao, Yunde Jia
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 U-Net refinement (ablation) | DepthSplat baseline · DL3DV 50 views | PSNR: 21.39 -> 21.71 | Improved |
| add history fusion (ablation) | LongSplat · DL3DV 50 views | PSNR: 21.71 -> 22.78 | Improved |
| add 3D dataset (reference 3DGS) supervision (ablation) | LongSplat · DL3DV 50 views | PSNR: 22.78 -> 23.12 | Improved |
| apply fixed compression mask on 3D dataset model (ablation) | LongSplat · DL3DV 50 views | PSNR: 23.12 -> 22.47 | Hurt |
| add learned compression fusion (ablation) | LongSplat · DL3DV 50 views | PSNR: 23.12 -> 23.71 | Improved |
| add adaptive masking to learned compression (ablation) | LongSplat · DL3DV 50 views | PSNR: 23.71 -> 23.54 with 43.77% fewer Gaussians | Mixed |
| increase mask threshold tau from none to 0.5 (ablation) | DepthSplat · DL3DV 50 views | PSNR: 21.39 -> 22.32 | Improved |
| increase mask threshold tau from none to 0.5 (ablation) | LongSplat · DL3DV 50 views | PSNR: 23.71 -> 23.54 at 43.77% compression | Mixed |
| LongSplat versus DepthSplat as view count grows to 120 | LongSplat · DL3DV 120 views | PSNR: 17.77 -> 21.02 | Improved |
| compress Gaussians with LongSplat-c at 12 views | LongSplat · DL3DV 12 views | PSNR: 22.68 -> 21.69 with 25.52% fewer Gaussians | Hurt |
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.
- DepthSplat — DepthSplat: Connecting Gaussian Splatting and Depth (CVPR 2025) · better on 6 of 6 shared results
- MVSplat-360 — MVSplat360: Feed-Forward 360 Scene Synthesis from Sparse Views (NeurIPS 2024) · 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
- HiCoM: Hierarchical Coherent Motion for Dynamic Streamable Scenes with 3D Gaussian Splatting (NeurIPS 2024 · similarity 0.92)
- GaussianSpa: An "Optimizing-Sparsifying" Simplification Framework for Compact and High-Quality 3D Gaussian Splatting (CVPR 2025 · similarity 0.91)
- Speedy-Splat: Fast 3D Gaussian Splatting with Sparse Pixels and Sparse Primitives (CVPR 2025 · similarity 0.91)
- SfM-Free 3D Gaussian Splatting via Hierarchical Training (CVPR 2025 · similarity 0.91)
- GaussianPro: 3D Gaussian Splatting with Progressive Propagation (ICML 2024 · similarity 0.91)
- Motion Matters: Compact Gaussian Streaming for Free-Viewpoint Video Reconstruction (NeurIPS 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.
