Research map / 3D vision / Gaussian splatting
StructSplat: Generalizable 3D Gaussian Splatting from Uncalibrated Sparse Views
Jia-Chen Zhao, Beiqi Chen, Xinyang Chen, Guangcong Wang, Liqiang Nie
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 semantic features to the geometric-only model (ablation) | StructSplat · DL3DV (2 source views) | PSNR: 20.610 -> 26.236 | Improved |
| add pixel-aligned texture features on top of geometry and semantics (ablation) | StructSplat · DL3DV (2 source views) | PSNR: 26.236 -> 28.045 | Improved |
| add semantic features to the geometric-only model (ablation) | StructSplat · DL3DV (2 source views) | LPIPS: 0.343 -> 0.151 | Improved |
| add camera alignment strategy (ablation) | StructSplat · DL3DV (2 source views) | PSNR: 27.338 -> 28.045 | Improved |
| add camera alignment strategy (ablation) | StructSplat · DL3DV (2 source views) | LPIPS: 0.097 -> 0.091 | Improved |
| use own camera alignment instead of AnySplat's translation-scale alignment | StructSplat · DL3DV | ATE position: 0.1003 -> 0.0093 | Improved |
| use own camera alignment instead of AnySplat's translation-scale alignment | StructSplat · DL3DV | ATE rotation: 0.0866 -> 0.0016 | Improved |
| apply pose post-optimization of target extrinsics | StructSplat · DL3DV | PSNR: 28.045 -> 29.287 | Improved |
| increase source views from 2 to 6 | StructSplat · DL3DV | PSNR: 28.045 -> 31.137 | Improved |
| sparser view sampling covering larger scenes (4 views) | StructSplat · DL3DV (larger-scale) | PSNR: +3.319 dB over AnySplat | Improved |
| evaluate at 448x448 resolution | StructSplat · DL3DV (high resolution) | PSNR: 27.073 vs 20.870 for AnySplat | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 4 on most shared results.
- RayZer — RayZer: A Self-supervised Large View Synthesis Model (ICCV 2025) · better on 3 of 3 shared results
- FLARE — FLARE: Feed-forward Geometry, Appearance and Camera Estimation from Uncalibrated Sparse Views (CVPR 2025) · better on 2 of 2 shared results
- PF3plat — PF3plat: Pose-Free Feed-Forward 3D Gaussian Splatting for Novel View Synthesis (ICML 2025) · better on 1 of 2 shared results
- SelfSplat — SelfSplat: Pose-Free and 3D Prior-Free Generalizable 3D Gaussian Splatting (CVPR 2025) · 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
- HiSplat: Hierarchical 3D Gaussian Splatting for Generalizable Sparse-View Reconstruction (ICLR 2025 · similarity 0.93)
- pixelSplat: 3D Gaussian Splats from Image Pairs for Scalable Generalizable 3D Reconstruction (CVPR 2024 · similarity 0.92)
- PF3plat: Pose-Free Feed-Forward 3D Gaussian Splatting for Novel View Synthesis (ICML 2025 · similarity 0.92)
- SelfSplat: Pose-Free and 3D Prior-Free Generalizable 3D Gaussian Splatting (CVPR 2025 · similarity 0.91)
- CompactSplat: Spatially Adaptive Gaussian Distribution for Feedforward Scene Reconstruction (NeurIPS 2026 · similarity 0.91)
- TranSplat: Generalizable 3D Gaussian Splatting from Sparse Multi-View Images with Transformers (AAAI 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.
