4K-NeRF: High Fidelity Neural Radiance Fields at Ultra High Resolutions

9 Dec 2022  ·  Zhongshu Wang, Lingzhi Li, Zhen Shen, Li Shen, Liefeng Bo ·

In this paper, we present a novel and effective framework, named 4K-NeRF, to pursue high fidelity view synthesis on the challenging scenarios of ultra high resolutions, building on the methodology of neural radiance fields (NeRF). The rendering procedure of NeRF-based methods typically relies on a pixel-wise manner in which rays (or pixels) are treated independently on both training and inference phases, limiting its representational ability on describing subtle details, especially when lifting to a extremely high resolution. We address the issue by exploring ray correlation to enhance high-frequency details recovery. Particularly, we use the 3D-aware encoder to model geometric information effectively in a lower resolution space and recover fine details through the 3D-aware decoder, conditioned on ray features and depths estimated by the encoder. Joint training with patch-based sampling further facilitates our method incorporating the supervision from perception oriented regularization beyond pixel-wise loss. Benefiting from the use of geometry-aware local context, our method can significantly boost rendering quality on high-frequency details compared with modern NeRF methods, and achieve the state-of-the-art visual quality on 4K ultra-high-resolution scenarios. Code Available at \url{https://github.com/frozoul/4K-NeRF}

PDF Abstract

Results from the Paper

  Submit results from this paper to get state-of-the-art GitHub badges and help the community compare results to other papers.