Global illumination with radiance regression functions

Ren, Peiran; Wang, Jiaping; Gong, Minmin; Lin, Stephen; Tong, Xin; Guo, Baining

doi:10.1145/2461912.2462009

Cited by 103 publications

(85 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Lastly, one of the promising approaches uses Precomputed Light Transport (PLT) for rendering complex scenes with a full indirect illumination in real-time and the ability to dynamically control local lights and viewing directions [112]. The approach uses radiance regression function to train the neural network in precomputation phase to allow real-time rendering of the indirect illumination of the whole scene under different lighting and viewing conditions.…”

Section: Types Of Simulations/analysesmentioning

confidence: 99%

State-of-the-art review of solar design tools and methods for assessing daylighting and solar potential for building-integrated photovoltaics

Jakica

2018

Renewable and Sustainable Energy Reviews

142

View full text Add to dashboard Cite

Solar design can take many different forms across disciplines with different methodologies and goals, ranging from acquiring architectural visual effects to assessing illumination for daylighting and solar radiation potential on building surfaces for PV implementation. Furthermore, a capability of solar design methodologies and tools to accurately and time efficiently simulate light phenomena can greatly influence performance results and design decisions. This is especially important in complex cases such as dense urban settings with the significant surface shadowing, and vertical facades including daylighting devices and photovoltaics. Consequently, choosing a suitable approach and tool for each design phase is essential for achieving unique design and performance goals. This paper was carried out within the framework of IEA-PVPS Task 15 -BIPV and it aims to facilitate this decision for all parties involved in solar design process. Here presented, is an overview of almost 200 solar design tools, analyzing their numerous features regarding accuracy, complexity, scale, computation speed, representation as well as building design process integration in about 50 2D/3D, CAD/CAM and BIM software environments. Furthermore, tools from various fields have been analysed in a broad interdisciplinary context of solar design with a particular attention for being used for Daylighting and Building-Integrated Photovoltaics (BIPV) purposes. This approach should open many new perspectives on a potentially wider multidisciplinary usage and interpretation of solar design tools, sometimes well beyond their initial scope of work.

show abstract

Section: Types Of Simulations/analysesmentioning

confidence: 99%

State-of-the-art review of solar design tools and methods for assessing daylighting and solar potential for building-integrated photovoltaics

Jakica

2018

Renewable and Sustainable Energy Reviews

142

View full text Add to dashboard Cite

show abstract

“…Ren et al [30] proposed Radiance Regression Functions which use an ANN to approximate the indirect illumination at every scene point in a computationally effective way. Their model is a nonlinear 6D radiance regression function which takes surface position, view direction and lighting direction as input and returns the RGB colour of the indirect illumination.…”

Section: Artificial Neural Networkmentioning

confidence: 99%

A Machine-Learning-Driven Sky Model

Satylmys

Bashford-Rogers

Chalmers

et al. 2017

IEEE Comput. Grap. Appl.

View full text Add to dashboard Cite

“…This limits the use cases to scenes where the irradiance can be precomputed. Ren et al describes an approach that uses radiance regression functions to compute global illumination using a neural network [41]. The underlying idea is similar to our approach but the method requires extensive pre-processing per scene and is limited to static geometry.…”

Section: Related Workmentioning

confidence: 99%

Real-Time Molecular Visualization Supporting Diffuse Interreflections and Ambient Occlusion

Skånberg

Vázquez

Guallar

et al. 2016

IEEE Trans. Visual. Comput. Graphics

View full text Add to dashboard Cite

(a) Time step 1 (b) Time step 2 (c) Time step 3 Fig. 1. By deriving analytic expressions, we can enhance molecular visualizations and realize interreflections in real-time. The images (a-c) show three time steps of a molecular simulation investigating the interaction between a magenta-colored ligand and a receptor molecule, which receives exaggerated diffuse interreflections. Due to these interreflections it can be seen how the ligand enters the active site.Abstract-Today molecular simulations produce complex data sets capturing the interactions of molecules in detail. Due to the complexity of this time-varying data, advanced visualization techniques are required to support its visual analysis. Current molecular visualization techniques utilize ambient occlusion as a global illumination approximation to improve spatial comprehension. Besides these shadow-like effects, interreflections are also known to improve the spatial comprehension of complex geometric structures. Unfortunately, the inherent computational complexity of interreflections would forbid interactive exploration, which is mandatory in many scenarios dealing with static and time-varying data. In this paper, we introduce a novel analytic approach for capturing interreflections of molecular structures in real-time. By exploiting the knowledge of the underlying space filling representations, we are able to reduce the required parameters and can thus apply symbolic regression to obtain an analytic expression for interreflections. We show how to obtain the data required for the symbolic regression analysis, and how to exploit our analytic solution to enhance interactive molecular visualizations.

show abstract

Global illumination with radiance regression functions

Cited by 103 publications

References 53 publications

State-of-the-art review of solar design tools and methods for assessing daylighting and solar potential for building-integrated photovoltaics

State-of-the-art review of solar design tools and methods for assessing daylighting and solar potential for building-integrated photovoltaics

A Machine-Learning-Driven Sky Model

Real-Time Molecular Visualization Supporting Diffuse Interreflections and Ambient Occlusion

Contact Info

Product

Resources

About