@inproceedings{95d856d4478e4818ac8e877ff6b92f00,
title = "Robust light fields denoising with S2N2N",
abstract = "We have recently introduced the Single Shot Noise2Noise (S2N2N) framework for denoising Light Fields (LF) captured by Integral Imaging (InI), which is robust against different types and intensities of noise at any arbitrary exposure. S2N2N implicitly learns the noise type and intensity from the captured LF. In this paper, we further investigate S2N2N and introduce several improvements, including integration with a Visual Image Transformer (ViT). We test the method using both synthetic and real-world datasets, demonstrating significant improvements in denoising performance, particularly in high-noise environments. Our results reveal that this unsupervised denoising method has significant potential for real-world 3D imaging applications, offering robust performance without the need for explicit noise models.",
keywords = "Deep Neural Networks, Denoise, Integral Imaging, Noise2Noise, SN2N, Three-Dimensional (3-D) Imaging, Visual Image Transformers",
author = "Tal Kozakov and Omer Hazan and Adir Hazan and Adrian Stern",
note = "Publisher Copyright: {\textcopyright} COPYRIGHT SPIE. Downloading of the abstract is permitted for personal use only.; Three-Dimensional Imaging, Visualization, and Display 2025 ; Conference date: 14-04-2025 Through 16-04-2025",
year = "2025",
month = jan,
day = "1",
doi = "10.1117/12.3053977",
language = "American English",
series = "Proceedings of SPIE - The International Society for Optical Engineering",
publisher = "SPIE",
editor = "Bahram Javidi and Xin Shen and Arun Anand",
booktitle = "Three-Dimensional Imaging, Visualization, and Display 2025",
address = "United States",
}