# Extreme View Synthesis

arXiv: https://arxiv.org/abs/1812.04777 · HTML: https://mlsanity.com/paper/1812.04777/

## Key facts

- Cited by 94 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 12 Dec 2018.

## Built on by (most cited first, 30 of 94)

- [MVSNeRF: Fast Generalizable Radiance Field Reconstruction from Multi-View Stereo](https://mlsanity.com/paper/2103.15595/) (2021)
- [DynIBaR: Neural Dynamic Image-Based Rendering](https://mlsanity.com/paper/2211.11082/) (2022)
- [NeX: Real-time View Synthesis with Neural Basis Expansion](https://mlsanity.com/paper/2103.05606/) (2021)
- [Robust Dynamic Radiance Fields](https://mlsanity.com/paper/2301.02239/) (2023)
- [Neural Rays for Occlusion-aware Image-based Rendering](https://mlsanity.com/paper/2107.13421/) (2021)
- [Stereo Radiance Fields (SRF): Learning View Synthesis for Sparse Views of Novel Scenes](https://mlsanity.com/paper/2104.06935/) (2021)
- [Point-Based Neural Rendering with Per-View Optimization](https://mlsanity.com/paper/2109.02369/) (2021)
- [LVSM: A Large View Synthesis Model with Minimal 3D Inductive Bias](https://mlsanity.com/paper/2410.17242/) (2024)
- [NeRF--: Neural Radiance Fields Without Known Camera Parameters](https://mlsanity.com/paper/2102.07064/) (2021)
- [Editable Free-viewpoint Video Using a Layered Neural Representation](https://mlsanity.com/paper/2104.14786/) (2021)
- [Unconstrained Scene Generation with Locally Conditioned Radiance Fields](https://mlsanity.com/paper/2104.00670/) (2021)
- [Hallucinated Neural Radiance Fields in the Wild](https://mlsanity.com/paper/2111.15246/) (2021)
- [HDR-NeRF: High Dynamic Range Neural Radiance Fields](https://mlsanity.com/paper/2111.14451/) (2021)
- [Light Field Neural Rendering](https://mlsanity.com/paper/2112.09687/) (2021)
- [Geometry-Free View Synthesis: Transformers and no 3D Priors](https://mlsanity.com/paper/2104.07652/) (2021)
- [NeRF-SR: High-Quality Neural Radiance Fields using Supersampling](https://mlsanity.com/paper/2112.01759/) (2021)
- [Neural Trajectory Fields for Dynamic Novel View Synthesis](https://mlsanity.com/paper/2105.05994/) (2021)
- [MonoHuman: Animatable Human Neural Field from Monocular Video](https://mlsanity.com/paper/2304.02001/) (2023)
- [InfiniteNature-Zero: Learning Perpetual View Generation of Natural Scenes from Single Images](https://mlsanity.com/paper/2207.11148/) (2022)
- [FoV-NeRF: Foveated Neural Radiance Fields for Virtual Reality](https://mlsanity.com/paper/2103.16365/) (2021)
- [NVS-Solver: Video Diffusion Model as Zero-Shot Novel View Synthesizer](https://mlsanity.com/paper/2405.15364/) (2024)
- [Neural Residual Radiance Fields for Streamably Free-Viewpoint Videos](https://mlsanity.com/paper/2304.04452/) (2023)
- [Fully Explicit Dynamic Gaussian Splatting](https://mlsanity.com/paper/2410.15629/) (2024)
- [ViewFormer: NeRF-free Neural Rendering from Few Images Using Transformers](https://mlsanity.com/paper/2203.10157/) (2022)
- [Forward Flow for Novel View Synthesis of Dynamic Scenes](https://mlsanity.com/paper/2309.17390/) (2023)
- [Advances in 3D Generation: A Survey](https://mlsanity.com/paper/2401.17807/) (2024)
- [Watch It Move: Unsupervised Discovery of 3D Joints for Re-Posing of Articulated Objects](https://mlsanity.com/paper/2112.11347/) (2021)
- [Persistent Nature: A Generative Model of Unbounded 3D Worlds](https://mlsanity.com/paper/2303.13515/) (2023)
- [Neural Deformable Voxel Grid for Fast Optimization of Dynamic View Synthesis](https://mlsanity.com/paper/2206.07698/) (2022)
- [FWD: Real-time Novel View Synthesis with Forward Warping and Depth](https://mlsanity.com/paper/2206.08355/) (2022)

## Abstract (arXiv)

We present Extreme View Synthesis, a solution for novel view extrapolation that works even when the number of input images is small--as few as two. In this context, occlusions and depth uncertainty are two of the most pressing issues, and worsen as the degree of extrapolation increases. We follow the traditional paradigm of performing depth-based warping and refinement, with a few key improvements. First, we estimate a depth probability volume, rather than just a single depth value for each pixel of the novel view. This allows us to leverage depth uncertainty in challenging regions, such as depth discontinuities. After using it to get an initial estimate of the novel view, we explicitly combine learned image priors and the depth uncertainty to synthesize a refined image with less artifacts. Our method is the first to show visually pleasing results for baseline magnifications of up to 30X.