Research map / 3D vision / 3D generation
Multi-view to Novel view: Synthesizing novel views with Self-Learned Confidence
Shao-Hua Sun, Minyoung Huh, Yuan-Hong Liao, Ning Zhang, Joseph J. Lim
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 the flow predictor to the pixel generator (ablation) | Pixel generator · ShapeNet Car (4 views) | L1: .082 -> .062 | Improved |
| add the pixel generator to the flow predictor (ablation) | Flow predictor · ShapeNet Car (4 views) | L1: .079 -> .062 | Improved |
| add the pixel generator to the flow predictor (ablation) | Flow predictor · ShapeNet Chair (4 views) | L1: .152 -> .111 | Improved |
| add the flow predictor to the pixel generator (ablation) | Pixel generator · ShapeNet Chair (4 views) | L1: .119 -> .111 | Improved |
| use self-learned confidence aggregation instead of an oracle pick of the best intermediate prediction (ablation) | Full model · ShapeNet Car (4 views) | L1: .070 -> .062 | Improved |
| use self-learned confidence aggregation instead of an oracle pick (ablation) | Full model · KITTI (2 views) | L1: .199 -> .163 | Improved |
| add the flow predictor to the pixel generator (ablation) | Pixel generator · KITTI (2 views) | L1: .234 -> .163 | Improved |
| add the pixel generator to the flow predictor (ablation) | Flow predictor · KITTI (1 view) | L1: .397 -> .203 | Improved |
| add the pixel generator to the flow predictor (ablation) | Flow predictor · Synthia (2 views) | SSIM: .720 -> .737 | Improved |
| use self-learned confidence aggregation instead of an oracle pick (ablation) | Full model · Synthia (2 views) | SSIM: .718 -> .737 | Improved |
| increase the number of source views from 1 to 4 | Ours · ShapeNet Car | L1: .098 -> .062 | Improved |
| increase the number of source views from 1 to 4 | Ours · ShapeNet Chair | L1: .181 -> .111 | Improved |
Closest papers by meaning
- Novel View Synthesis with Pixel-Space Diffusion Models (CVPR 2025 · similarity 0.89)
- NAMVIS: Next-Scale Autoregressive Multi-View Image Synthesis (NeurIPS 2026 · similarity 0.89)
- UMAMI: Unifying Masked Autoregressive Models and Deterministic Rendering for View Synthesis (NeurIPS 2025 · similarity 0.89)
- One View Is Enough: In-the-Wild Monocular Pretraining for Novel View Generation (NeurIPS 2026 · similarity 0.88)
- Efficient-3Dim: Learning a Generalizable Single-image Novel-view Synthesizer in One Day (ICLR 2024 · similarity 0.88)
- DreamComposer: Controllable 3D Object Generation via Multi-View Conditions (CVPR 2024 · 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.
