Random asymptotic photon mapping image denoising method and system based on neural network

A neural network and photon mapping technology, used in image enhancement, image analysis, image data processing, etc., can solve the problem of increasing the difficulty of noise reduction, and achieve the effect of increasing stability, maintaining lighting details, and ensuring quality.

Active Publication Date: 2020-07-24
SHANDONG UNIV
View PDF7 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The special nature of random progressive photon mapping introduces multi-scale nois

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Random asymptotic photon mapping image denoising method and system based on neural network
  • Random asymptotic photon mapping image denoising method and system based on neural network
  • Random asymptotic photon mapping image denoising method and system based on neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0039] Embodiment 1, this embodiment provides a neural network-based random asymptotic photon mapping image noise reduction method;

[0040] A random asymptotic photon mapping image noise reduction method based on neural network, including:

[0041] S100: Acquire a three-dimensional scene, and generate an image to be denoised based on random progressive photon mapping, where the image to be denoised includes a rendered image based on global photons and a rendered image based on caustic photons;

[0042] S200: Input the rendered image based on global photons into the pre-trained first multiple residual neural network, and output the global photon noise reduction image; input the rendered image based on caustic photons into the pre-trained second multiple In the residual neural network, the caustic photon noise reduction image is output;

[0043] S300: Combining the global photon noise reduction image and the caustic photon noise reduction image to obtain a final rendered image...

Embodiment 2

[0116] Embodiment 2, this embodiment provides a neural network-based random asymptotic photon mapping image noise reduction system;

[0117] A neural network based stochastic asymptotic photon mapping image denoising system, including:

[0118] A data generation module configured to: acquire a three-dimensional scene, generate an image to be denoised based on random progressive photon mapping, the image to be denoised includes a rendered image based on global photons and a rendered image based on caustic photons;

[0119] The noise reduction module is configured to: input the rendered image based on the global photon into the pre-trained first multiple residual neural network, and output the global photon noise reduction image; input the rendered image based on the caustic photon into In the pre-trained second multiple residual neural network, the caustic photon noise reduction image is output;

[0120] The synthesis module is configured to: synthesize the global photon noise...

Embodiment 3

[0122] Embodiment 3. This embodiment also provides an electronic device, including a memory, a processor, and computer instructions stored in the memory and run on the processor. When the computer instructions are executed by the processor, the computer instructions in Embodiment 1 are completed. described method.

[0123] It should be understood that in this embodiment, the processor can be a central processing unit CPU, and the processor can also be other general-purpose processors, digital signal processors DSP, application specific integrated circuits ASIC, off-the-shelf programmable gate array FPGA or other programmable logic devices , discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor, or the processor may be any conventional processor, or the like.

[0124] The memory may include read-only memory and random access memory, and provides instructions and data to the processor, and a part of the...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a random asymptotic photon mapping image denoising method and system based on a neural network, and the method comprises the steps: obtaining a three-dimensional scene, and generating a to-be-denoised image based on random asymptotic photon mapping, wherein the to-be-denoised image comprises a rendering image based on global photons and a rendering image based on focal astigmatism; inputting the rendering image based on the global photons into a pre-trained first multiple residual neural network, and outputting a global photon denoising image; inputting the rendered image based on the focal astigmatism photons into a pre-trained second multiple residual neural network, and outputting a focal astigmatism photon denoising image; and synthesizing the global photon denoising image and the focal astigmatism photon denoising image to obtain a final rendering image.

Description

technical field [0001] The present disclosure relates to the technical field of image realistic rendering, in particular to a neural network-based random asymptotic photon mapping image noise reduction method and system. Background technique [0002] The statements in this section merely mention background art related to the present disclosure and do not necessarily constitute prior art. [0003] Random progressive photon mapping is a widely used physically-based realistic rendering method, which can effectively simulate complex optical effects such as caustics produced by the interaction between light and objects in the real world. It is a memory-friendly photon mapping and improved methods for progressive photon mapping. [0004] However, as a biased rendering method, stochastic progressive photon mapping often produces low-quality results due to large timing bias and variance when inappropriate scene rendering parameters are selected or rendering iterations are insuffici...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
IPC IPC(8): G06T5/00G06T15/00
CPCG06T5/002G06T15/00G06T2207/20081G06T2207/20084
Inventor 王璐曾峥徐延宁康春萌王贝贝孟祥旭
Owner SHANDONG UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap