Research map / 3D vision / Gaussian splatting
GaussianZoom: Progressive Zoom-in Generative 3D Gaussian Splatting with Geometric and Semantic Guidance
Jiale Shi, Jiarui Hu, Zesong Yang, Kaixuan Luan, Hujun Bao, Zhaopeng Cui
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 depth-guided feature warping (replacing flow-based warping) (ablation) | GaussianZoom · Mip-NeRF360 | FVD: 168.36 -> 107.99 | Improved |
| add depth-guided feature warping (replacing flow-based warping) (ablation) | GaussianZoom · Tanks&Temples | FVD: 180.45 -> 79.98 | Improved |
| add VLM-inferred semantic prompts to the SR model (ablation) | GaussianZoom · Tanks&Temples truck (qualitative) | semantic/material consistency | Improved |
| add continuous LoD hierarchy with separate Gaussian layers per scale (ablation) | GaussianZoom · zoom-in and zoom-out renderings (qualitative) | aliasing / cross-scale consistency | Improved |
| GaussianZoom vs Sequence Matters (best baseline) at 4x SR | GaussianZoom · Mip-NeRF360 4x SR | FID: 26.64 -> 19.38 | Improved |
| GaussianZoom vs Sequence Matters at 4x SR | GaussianZoom · Tanks&Temples 4x SR | PSNR: 23.39 -> 23.40 | No clear effect |
| GaussianZoom vs SRGS at 64x extreme zoom-in | GaussianZoom · Extreme zoom-in 64x | MUSIQ: 17.27 -> 42.21 | Improved |
| GaussianZoom vs Sequence Matters at 64x extreme zoom-in | GaussianZoom · Extreme zoom-in 64x | NIQE: 15.25 -> 5.53 | Improved |
| GaussianZoom vs SRGS at 16x extreme zoom-in | GaussianZoom · Extreme zoom-in 16x | CLIPIQA: 0.272 -> 0.347 | Improved |
Closest papers by meaning
- Generative Sparse-View Gaussian Splatting (CVPR 2025 · similarity 0.90)
- PUP 3D-GS: Principled Uncertainty Pruning for 3D Gaussian Splatting (CVPR 2025 · similarity 0.89)
- VastGaussian: Vast 3D Gaussians for Large Scene Reconstruction (CVPR 2024 · similarity 0.89)
- CompactSplat: Spatially Adaptive Gaussian Distribution for Feedforward Scene Reconstruction (NeurIPS 2026 · similarity 0.89)
- GaussianPro: 3D Gaussian Splatting with Progressive Propagation (ICML 2024 · similarity 0.89)
- Generative Densification: Learning to Densify Gaussians for High-Fidelity Generalizable 3D Reconstruction (CVPR 2025 · similarity 0.89)
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.
