Research map / 3D vision / 3D generation
Auto-Regressively Generating Multi-View Consistent Images
JiaKui Hu, Yuxiao Yang, Jialun Liu, Jinbo Wu, Chen Zhao, Yanye Lu
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 Shuffle Views augmentation (ablation) | MV-AR · GSO text-to-multi-view | FID: 173.51 -> 144.29 | Improved |
| add Shuffle Views augmentation (ablation) | MV-AR · GSO text-to-multi-view | IS: 4.77 -> 8.00 | Improved |
| add Shuffle Views augmentation (ablation) | MV-AR · GSO image-to-multi-view | PSNR: 18.27 -> 22.99 | Improved |
| add shift position embedding from camera pose (ablation) | MV-AR · GSO text-to-multi-view | FID: 147.29 -> 144.29 | Improved |
| add shift position embedding from camera pose (ablation) | MV-AR · GSO image-to-multi-view | PSNR: 21.30 -> 22.99 | Improved |
| use Image Warp Controller instead of in-context image condition (ablation) | MV-AR · GSO image-to-multi-view | PSNR: 11.92 -> 22.99 | Improved |
| use Image Warp Controller instead of cross attention (ablation) | MV-AR · GSO image-to-multi-view | LPIPS: 0.310 -> 0.084 | Improved |
| use Split Self-Attention for text condition instead of plain self-attention (ablation) | MV-AR · GSO text-to-multi-view | CLIP-Score: 28.36 -> 29.49 | Improved |
| use Split Self-Attention for text condition instead of plain self-attention (ablation) | MV-AR · GSO text-to-multi-view | IS: 5.78 -> 8.00 | Improved |
| use Split Self-Attention (MV-AR) instead of plain self-attention (ablation) | MV-AR · GSO text-to-multi-view | FID: 146.11 -> 144.29 | Improved |
| auto-regressive MV-AR instead of diffusion MVDream | MV-AR · GSO text-to-multi-view | FID: 141.05 -> 144.29 | Hurt |
| auto-regressive MV-AR instead of diffusion Wonder3D | MV-AR · GSO image-to-multi-view | LPIPS: 0.062 -> 0.084 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- SyncDreamer — SyncDreamer: Generating Multiview-consistent Images from a Single-view Image (ICLR 2024) · better on 3 of 3 shared results
- RealFusion — RealFusion: 360deg Reconstruction of Any Object From a Single Image (CVPR 2023) · better on 3 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
- MVGenMaster: Scaling Multi-View Generation from Any Image via 3D Priors Enhanced Diffusion Model (CVPR 2025 · similarity 0.88)
- NAMVIS: Next-Scale Autoregressive Multi-View Image Synthesis (NeurIPS 2026 · similarity 0.88)
- MVDream: Multi-view Diffusion for 3D Generation (ICLR 2024 · similarity 0.88)
- ViewDiff: 3D-Consistent Image Generation with Text-to-Image Models (CVPR 2024 · similarity 0.87)
- MVD-Fusion: Single-view 3D via Depth-consistent Multi-view Generation (CVPR 2024 · similarity 0.86)
- Vivid-ZOO: Multi-View Video Generation with Diffusion Model (NeurIPS 2024 · similarity 0.86)
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.
