Research map / 3D vision / Gaussian splatting
SV-GS: Sparse View 4D Reconstruction with Skeleton-Driven Gaussian Splatting
Jun-Jee Chao, Volkan Isler
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 detail deformation field MLP_Psi (ablation) | SV-GS · D-NeRF (11 frames) | PSNR: 26.34 -> 27.75 | Improved |
| add skinning correction field MLP_Phi (ablation) | SV-GS · D-NeRF (11 frames) | PSNR: 27.28 -> 27.75 | Improved |
| add motion regularization L_motion (ablation) | SV-GS · D-NeRF (11 frames) | PSNR: 27.26 -> 27.75 | Improved |
| add detail deformation field MLP_Psi (ablation) | SV-GS · D-NeRF (11 frames) | LPIPS (x100): 6.51 -> 5.79 | Improved |
| use skeleton-driven deformation with time-dependent joint poses instead of 4DGS-style deformation field | sparse-view 4D reconstruction · D-NeRF (11 frames) | PSNR: 21.70 -> 27.75 | Improved |
| give RigGS the same skeleton input as SV-GS | RigGS · D-NeRF (11 frames) | PSNR: 24.23 -> 22.63 | Hurt |
| increase temporal observations from 11 to 21 frames | baselines on DG-Mesh · DG-Mesh SSIM | SSIM: 4DGS 0.887 -> 0.918 | Improved |
| use 1/5 instead of 1/10 of frames for SV-GS | SV-GS · ZJU-MoCap | PSNR: 28.13 -> 28.83 | Improved |
| sparse-frame SV-GS vs RigGS trained on full monocular video | SV-GS · ZJU-MoCap | PSNR: 33.54 -> 28.83 | Hurt |
| replace multi-view initial reconstruction with Zero-1-to-3 SDS prior from one image | SV-GS · D-NeRF Jumpingjacks, DAVIS camel | qualitative | Mixed |
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.
- RigGS — RigGS: Rigging of 3D Gaussians for Modeling Articulated Objects in Videos (CVPR 2025) · better on 3 of 3 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
- Motion Decoupled 3D Gaussian Splatting for Dynamic Object Representation (AAAI 2025 · similarity 0.89)
- DGS-LRM: Real-Time Deformable 3D Gaussian Reconstruction From Monocular Videos (NeurIPS 2025 · similarity 0.88)
- SC-GS: Sparse-Controlled Gaussian Splatting for Editable Dynamic Scenes (CVPR 2024 · similarity 0.88)
- 4D3R: Motion-Aware Neural Reconstruction and Rendering of Dynamic Scenes from Monocular Videos (NeurIPS 2025 · similarity 0.87)
- MoDec-GS: Global-to-Local Motion Decomposition and Temporal Interval Adjustment for Compact Dynamic 3D Gaussian Splatting (CVPR 2025 · similarity 0.87)
- S3O: A Dual-Phase Approach for Reconstructing Dynamic Shape and Skeleton of Articulated Objects from Single Monocular Video (ICML 2024 · similarity 0.87)
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.
