Research map / Video / Video segmentation and tracking
SuperGaussian: Repurposing Video Models for 3D Super Resolution
Yuan Shen, Duygu Ceylan, Paul Guerrero, Zexiang Xu, Niloy J. Mitra, Shenlong Wang, Anna Fruehstueck
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 |
|---|---|---|---|
| use video upsampler prior instead of per-frame image upsampler in 3D upsampling (ablation) | SuperGaussian · MVImgNet | LPIPS: 0.1524 -> 0.1290 (pre-hoc image vs ours) | Improved |
| use video upsampler instead of image upsampler on frames (ablation) | upsampler prior · MVImgNet upsampled frames | NIQE: 7.06 -> 6.04 | Improved |
| video upsampler instead of image upsampler on frames (ablation) | upsampler prior · MVImgNet upsampled frames | LPIPS: 0.1533 -> 0.1511 | No clear effect |
| upsample all frames once (pre-hoc) instead of iterative image-upsampler optimization (ablation) | image-based baselines · MVImgNet | LPIPS: 0.1867 -> 0.1524 (Instruct-G2G vs pre-hoc) | Improved |
| use Gaussian Splats as output representation instead of Nerfacto hashgrid or triplane (ablation) | SuperGaussian · chestnut scene | LPIPS: 0.3495 -> 0.1493 | Improved |
| use video prior without finetuning on unseen scene type (ablation) | SuperGaussian · Wild RGB-D boat | LPIPS: 0.257 -> 0.242 | Improved |
| sample upsampling trajectory closer to the target (ablation) | SuperGaussian · trajectory ablation | qualitative | Improved |
| SuperGaussian vs FastSR-NeRF on Blender synthetic | SuperGaussian · Blender synthetic (x4) | PSNR: 30.47 -> 28.44 | Hurt |
| SuperGaussian vs NeRF-SR and FastSR-NeRF on Blender synthetic | SuperGaussian · Blender synthetic (x4) | LPIPS: 0.075 -> 0.067 | Improved |
Closest papers by meaning
- S2Gaussian: Sparse-View Super-Resolution 3D Gaussian Splatting (CVPR 2025 · similarity 0.90)
- MaskGaussian: Adaptive 3D Gaussian Representation from Probabilistic Masks (CVPR 2025 · similarity 0.89)
- Sequence Matters: Harnessing Video Models in 3D Super-Resolution (AAAI 2025 · similarity 0.88)
- LightGaussian: Unbounded 3D Gaussian Compression with 15x Reduction and 200+ FPS (NeurIPS 2024 · similarity 0.88)
- Generative Densification: Learning to Densify Gaussians for High-Fidelity Generalizable 3D Reconstruction (CVPR 2025 · similarity 0.88)
- DropGaussian: Structural Regularization for Sparse-view Gaussian Splatting (CVPR 2025 · similarity 0.88)
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.
