Research map / 3D vision / 3D generation
Viewpoint textual inversion: discovering scene representations and 3D view control in 2D diffusion models
James Burgess, Kuan-Chieh Wang, Serena Yeung-Levy
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 ViewNeTI instead of prior NVS baselines for single-image view synthesis | ViewNeTI · DTU single-image NVS | LPIPS: 0.421 -> 0.378 (vs NerDi) | Improved |
| use ViewNeTI instead of pixelNeRF or SinNeRF on pixel-fidelity metrics | ViewNeTI · DTU single-image NVS | PSNR: 16.520 (SinNeRF) vs 10.947 | Hurt |
| add multi-scene pretraining of the view mapper (ablation) | ViewNeTI view token · view generalization | extrapolation and cross-scene generalization (qualitative) | Improved |
| remove image augmentations in single-scene training (ablation) | ViewNeTI · single-scene optimization (qualitative) | scene semantics learning | Improved |
| use too-low frequency Fourier encoding for camera parameters (ablation) | ViewNeTI · single-scene optimization (qualitative) | view coverage | Hurt |
| train ViewNeTI on 3-9 posed images of one scene, evaluate extrapolated views (ablation) | single-scene ViewNeTI · DTU scene 114 | view synthesis (qualitative) | Mixed |
| use ViewNeTI instead of Zero-1-to-3 trained on 800,000 scenes | ViewNeTI · DTU NVS (small pretraining) | LPIPS: 0.40 -> 0.37 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- SinNeRF — SinNeRF: Training Neural Radiance Fields on Complex Scenes from a Single Image (ECCV 2022) · better on 1 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
- IT3D: Improved Text-to-3D Generation with Explicit View Synthesis (AAAI 2024 · similarity 0.89)
- SweetDreamer: Aligning Geometric Priors in 2D diffusion for Consistent Text-to-3D (ICLR 2024 · similarity 0.88)
- Let 2D Diffusion Model Know 3D-Consistency for Robust Text-to-3D Generation (ICLR 2024 · similarity 0.88)
- A 3D Scene is Worth 1K Tokens: 3D-Grounded Representation for Scene Generation at Scale (NeurIPS 2026 · similarity 0.88)
- MVDream: Multi-view Diffusion for 3D Generation (ICLR 2024 · similarity 0.87)
- ViewDiff: 3D-Consistent Image Generation with Text-to-Image Models (CVPR 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.
