Deep Adaptive Sampling and Reconstruction Using Analytic Distributions

Author:

Salehi Farnood1,Manzi Marco1,Roethlin Gerhard1,Weber Romann1,Schroers Christopher1,Papas Marios1

Affiliation:

1. Disney Research | Studios, Switzerland

Abstract

We propose an adaptive sampling and reconstruction method for offline Monte Carlo rendering. Our method produces sampling maps constrained by a user-defined budget that minimize the expected future denoising error. Compared to other state-of-the-art methods, which produce the necessary training data on the fly by composing pre-rendered images, our method samples from analytic noise distributions instead. These distributions are compact and closely approximate the pixel value distributions stemming from Monte Carlo rendering. Our method can efficiently sample training data by leveraging only a few per-pixel statistics of the target distribution, which provides several benefits over the current state of the art. Most notably, our analytic distributions' modeling accuracy and sampling efficiency increase with sample count, essential for high-quality offline rendering. Although our distributions are approximate, our method supports joint end-to-end training of the sampling and denoising networks. Finally, we propose the addition of a global summary module to our architecture that accumulates valuable information from image regions outside of the network's receptive field. This information discourages sub-optimal decisions based on local information. Our evaluation against other state-of-the-art neural sampling methods demonstrates denoising quality and data efficiency improvements.

Publisher

Association for Computing Machinery (ACM)

Subject

Computer Graphics and Computer-Aided Design

Reference35 articles.

1. Martín Abadi Ashish Agarwal Paul Barham etal 2015. TensorFlow: Large-scale machine learning on heterogeneous systems. http://tensorflow.org/ Martín Abadi Ashish Agarwal Paul Barham et al. 2015. TensorFlow: Large-scale machine learning on heterogeneous systems. http://tensorflow.org/

2. FLIP

3. Kernel-predicting convolutional networks for denoising Monte Carlo renderings

4. Combining edges and points for interactive high-quality rendering

5. General and Robust Error Estimation and Reconstruction for Monte Carlo Rendering. Computer Graphics Forum (Proc;Bauszat Pablo;of Eurographics EG),2015

Cited by 2 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

1. Input-Dependent Uncorrelated Weighting for Monte Carlo Denoising;SIGGRAPH Asia 2023 Conference Papers;2023-12-10

2. Perspectives and Final Remarks;SpringerBriefs in Mathematics;2023

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3