cryptowealth-technology / r2l Goto Github PK

This repository is for the new Neral Light Field (NeLF) method introduced in the following paper:

Distilling Neural Radiance Field to Neural Light Field for Efficient Novel View Synthesis [Arxiv] [Project]
Huan Wang [1,2], Jian Ren [1], Zeng Huang [1], Kyle Olszewski [1], Menglei Chai [1], Yun Fu [2], and Sergey Tulyakov [1]
[1] Snap Inc., [2] Northeastern University,
Work done when Huan was an intern at Snap Inc.

[TL;DR] We present R2L, a deep (88-layer) residual MLP network that can represent the neural light field (NeLF) of complex synthetic and real-world scenes. It is featured by compact representation size (~20MB storage size), faster rendering speed (~30x speedup than NeRF), significantly improved visual quality (1.4dB boost than NeRF), with no whistles and bells (no special data structure or parallelism required).

Code will be released soon. Stay tuned!

See our project webpage