Research map / 3D vision / Gaussian splatting
Real-Time Dynamic Scene Rendering with Controlled Compressibility and Contact Awareness
Boya Shi, Naiyang Guan, Xiaodong Yi
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 source-aware compressible flow only (ablation) | Source-aware compressible Gaussian rendering · D-NeRF ablation | PSNR: 28.97 -> 32.50 | Improved |
| add source-aware compressible flow only (ablation) | Source-aware compressible Gaussian rendering · D-NeRF ablation | LPIPS: 0.19 -> 0.05 | Improved |
| add manifold and contact constraints only (ablation) | Source-aware compressible Gaussian rendering · D-NeRF ablation | PSNR: 28.97 -> 30.30 | Improved |
| add manifold and contact constraints only (ablation) | Source-aware compressible Gaussian rendering · D-NeRF ablation | LPIPS: 0.19 -> 0.07 | Improved |
| combine compressible flow and contact constraints (ablation) | Source-aware compressible Gaussian rendering · D-NeRF ablation | PSNR: 32.50 -> 34.80 | Improved |
| add contact constraints on top of compressible flow (ablation) | Source-aware compressible Gaussian rendering · D-NeRF ablation | training time: 16 min -> 13 min | Improved |
| use local-linear projection instead of Gaussian-primitive projection (ablation) | Source-aware compressible Gaussian rendering · D-NeRF ablation | PSNR: 35.00 -> 35.24 | Improved |
| use local-linear projection instead of Gaussian-primitive projection (ablation) | Source-aware compressible Gaussian rendering · D-NeRF ablation | training time: 8 min -> 6 min | Improved |
| use Ours instead of strongest prior on Plenoptic Video | Source-aware compressible Gaussian rendering · Plenoptic Video Dataset | PSNR: 30.85 -> 33.84 | Improved |
| use Ours instead of Deformable3DGS on D-NeRF | Source-aware compressible Gaussian rendering · D-NeRF | PSNR: 39.31 -> 35.24 | Hurt |
| use Ours instead of Deformable4DGS on D-NeRF | Source-aware compressible Gaussian rendering · D-NeRF | PSNR: 32.99 -> 35.24 | Improved |
Closest papers by meaning
- Real-time Photorealistic Dynamic Scene Representation and Rendering with 4D Gaussian Splatting (ICLR 2024 · similarity 0.89)
- FreeTimeGS: Free Gaussian Primitives at Anytime Anywhere for Dynamic Scene Reconstruction (CVPR 2025 · similarity 0.88)
- DySurface: Consistent 4D Surface Reconstruction via Bridging Explicit Gaussians and Implicit Functions (NeurIPS 2026 · similarity 0.88)
- ReCon-GS: Continuum-Preserved Guassian Streaming for Fast and Compact Reconstruction of Dynamic Scenes (NeurIPS 2025 · similarity 0.87)
- Spacetime Gaussian Feature Splatting for Real-Time Dynamic View Synthesis (CVPR 2024 · similarity 0.87)
- Fully Explicit Dynamic Gaussian Splatting (NeurIPS 2024 · similarity 0.86)
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.
