Research map / 3D vision / 3D generation
Vista3D: unravel the 3d darkside of a single image
Qiuhong Shen, Xingyi Yang, Michael Bi Mi, Xinchao Wang
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 second diffusion prior (Stable Diffusion) with angular score composition (Vista3D-L versus Vista3D-S) (ablation) | Vista3D · RealFusion dataset | CLIP-Similarity: 0.831 -> 0.868 | Improved |
| add angular prior composition (Vista3D-L versus Vista3D-S) (ablation) | Vista3D · GSO (30 objects) | PSNR: 25.42 -> 26.31 | Improved |
| use naive fixed weight 1/40 on the 2D diffusion SDS term instead of angular composition (ablation) | Vista3D-L · qualitative examples | back-view texture | Hurt |
| remove angular score composition (ablation) | Vista3D-L · qualitative examples | front-view consistency with reference | Hurt |
| skip coarse Gaussian-splatting stage and learn isosurface geometry directly (ablation) | Vista3D · qualitative examples | geometry quality | Hurt |
| use two disentangled hash encodings (forward and back) instead of one (ablation) | Vista3D · qualitative examples | texture artifacts | Improved |
| use Top-K gradient-based densification instead of naive gradient threshold (ablation) | Vista3D coarse stage · qualitative examples | geometry quality | Improved |
| add transmittance regularization to the coarse stage (ablation) | Vista3D coarse stage · qualitative examples | mesh watertightness | Improved |
| add scale regularization to the coarse stage (ablation) | Vista3D coarse stage · qualitative examples | geometry detail | Improved |
| use Vista3D-S instead of Magic123 for image-to-3D | image-to-3D generation · RealFusion dataset | CLIP-Similarity and time: 0.802 -> 0.831 | 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.
- Magic123 — Magic123: One Image to High-Quality 3D Object Generation Using Both 2D and 3D Diffusion Priors (ICLR 2024) · better on 5 of 6 shared results
- DreamGaussian — DreamGaussian: Generative Gaussian Splatting for Efficient 3D Content Creation (ICLR 2024) · better on 3 of 3 shared results
- DreamCraft3D — DreamCraft3D: Hierarchical 3D Generation with Bootstrapped Diffusion Prior (ICLR 2024) · better on 1 of 2 shared results
- Zero-1-to-3 — Zero-1-to-3: Zero-shot One Image to 3D Object (ICCV 2023) · 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
- UVGS: Reimagining Unstructured 3D Gaussian Splatting using UV Mapping (CVPR 2025 · similarity 0.89)
- 3D Convex Splatting: Radiance Field Rendering with 3D Smooth Convexes (CVPR 2025 · similarity 0.88)
- VA-GS: Enhancing the Geometric Representation of Gaussian Splatting via View Alignment (NeurIPS 2025 · similarity 0.88)
- Denoising Diffusion via Image-Based Rendering (ICLR 2024 · similarity 0.88)
- Neural Signed Distance Function Inference through Splatting 3D Gaussians Pulled on Zero-Level Set (NeurIPS 2024 · similarity 0.88)
- FruitNinja: 3D Object Interior Texture Generation with 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.
