Research map / 3D vision / 3D generation
Geometrically Consistent Multi-View Scene Generation from Freehand Sketches
Ahmed Bourouis, Savas Ozkan, Andrea Maracani, Yi-Zhe Song, Mete Ozay
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 correspondence supervision loss (CSL) (ablation) | S2MV · S2MV test set (N=33 views) | Corr-Acc: 0.175 -> 0.199 | Improved |
| add correspondence supervision loss (CSL) (ablation) | S2MV · S2MV test set (N=33 views) | FID: 20.37 -> 18.49 | Improved |
| add correspondence supervision loss (CSL) (ablation) | S2MV · S2MV test set (N=33 views) | LPIPS: 0.664 -> 0.632 | Improved |
| add CA3 camera adapters (ablation) | S2MV (LoRA only) · S2MV test set (N=33 views) | FID: 266.06 -> 18.49 | Improved |
| add CA3 camera adapters (ablation) | S2MV (LoRA only) · S2MV test set (N=33 views) | PSNR: 5.026 -> 12.169 | Improved |
| add CA3 camera adapters (ablation) | S2MV (LoRA only) · S2MV test set (N=33 views) | CLIP-I: 0.632 -> 0.828 | Improved |
| add LoRA domain adaptation (ablation) | S2MV (CA3 only) · S2MV test set (N=33 views) | Corr-Acc: 0.188 -> 0.199 | Improved |
| add LoRA domain adaptation (ablation) | S2MV (CA3 only) · S2MV test set (N=33 views) | PSNR: 12.211 -> 12.169 | Hurt |
| add frame replication (N -> 4N+1) (ablation) | S2MV · S2MV test set (N=33 views) | FID: 42.54 -> 18.49 | Improved |
| add frame replication (N -> 4N+1) (ablation) | S2MV · S2MV test set (N=33 views) | Corr-Acc: 0.174 -> 0.199 | Improved |
| replace two-stage pipeline with single-stage sketch-to-multiview | S2MV vs SEVA · S2MV test set (N=33 views) | FID: 46.34 -> 18.49 | Improved |
| replace two-stage pipeline with single-stage sketch-to-multiview | S2MV vs ViewCrafter · S2MV test set (N=33 views) | SSIM: 0.338 -> 0.302 | Hurt |
Closest papers by meaning
- MultiDiff: Consistent Novel View Synthesis from a Single Image (CVPR 2024 · similarity 0.88)
- MET3R: Measuring Multi-View Consistency in Generated Images (CVPR 2025 · similarity 0.87)
- MVInpainter: Learning Multi-View Consistent Inpainting to Bridge 2D and 3D Editing (NeurIPS 2024 · similarity 0.87)
- NAMVIS: Next-Scale Autoregressive Multi-View Image Synthesis (NeurIPS 2026 · similarity 0.87)
- Ouroboros3D: Image-to-3D Generation via 3D-aware Recursive Diffusion (CVPR 2025 · similarity 0.87)
- GenRec: Knowing Where to Reconstruct and Where to Generate (NeurIPS 2026 · 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.
