arXiv (Cornell University)· 2020· Preprint
NeRF++: Analyzing and Improving Neural Radiance Fields
- 510citations
- 2020year
Short summary
NeRF++ introduces a novel parameterization to improve Neural Radiance Fields (NeRF) for synthesizing novel views in large-scale, unbounded 3D scenes, addressing a key limitation of the original NeRF method.
AI-generated from the title and abstract; the full text is not read.
Key points
- NeRF++ addresses the shape-radiance ambiguity inherent in radiance fields.
- A new parameterization is introduced to improve NeRF's application to 360 captures of objects in large, unbounded scenes.
- The method enhances view synthesis fidelity in complex, large-scale environments.
- NeRF uses MLPs to represent scene opacity and color for novel view generation via volume rendering.
AI-generated from the title and abstract; the full text is not read.
Abstract
Neural Radiance Fields (NeRF) achieve impressive view synthesis results for a variety of capture settings, including 360 capture of bounded scenes and forward-facing capture of bounded and unbounded scenes. NeRF fits multi-layer perceptrons (MLPs) representing view-invariant opacity and view-dependent color volumes to a set of training images, and samples novel views based on volume rendering techniques. In this technical report, we first remark on radiance fields and their potential ambiguities, namely the shape-radiance ambiguity, and analyze NeRF's success in avoiding such ambiguities. Second, we address a parametrization issue involved in applying NeRF to 360 captures of objects within large-scale, unbounded 3D scenes. Our method improves view synthesis fidelity in this challenging scenario. Code is available at https://github.com/Kai-46/nerfplusplus.
The authors' abstract, as published at the source. arXiv (Cornell University), 2020 · DOI ↗
Continue with a free account
Ask the paper: 3 free questions a day about this paper; save it, get its citation, new summaries every day for your field. Takeaways are Premium.
Continue free on the webSign in with Google or Apple; no card needed. You come back to this paper.
On your phone:
Field: Computer Graphics and Computer-Aided Design
Computer Graphics and Computer-Aided DesignComputer Science