Topic
Radiance
About: Radiance is a research topic. Over the lifetime, 9537 publications have been published within this topic receiving 215652 citations.
Papers published on a yearly basis
Papers
More filters
TL;DR: Zhang et al. as discussed by the authors proposed a 3D-aware portrait lighting control model by distilling the shading from the original fused representation of both appearance and lighting with the conditional discriminator to supervise the lighting effects.
Abstract: 3D-aware portrait lighting control is an emerging and promising domain, thanks to the recent advance of generative adversarial networks and neural radiance fields. Existing solutions typically try to decouple the lighting from the geometry and appearance for disentangled control with an explicit lighting representation (e.g., Lambertian or Phong). However, they either are limited to a constrained lighting condition (e.g., directional light) or demand a tricky-to-fetch dataset as supervision for the intrinsic compositions (e.g., the albedo). We propose NeRFFaceLighting to explore an implicit representation for portrait lighting based on the pretrained tri-plane representation to address the above limitations. We approach this disentangled lighting-control problem by distilling the shading from the original fused representation of both appearance and lighting (i.e., one tri-plane) to their disentangled representations (i.e., two tri-planes) with the conditional discriminator to supervise the lighting effects. We further carefully design the regularization to reduce the ambiguity of such decomposition and enhance the ability of generalization to unseen lighting conditions. Moreover, our method can be extended to enable 3D-aware real portrait relighting. Through extensive quantitative and qualitative evaluations, we demonstrate the superior 3D-aware lighting control ability of our model compared to alternative and existing solutions.
TL;DR: In this paper , a radiative transfer model, libRadtran 2.0.4, was used to calculate UV-A radiation and compare it with observed data in Daejeon as the true value for the cloud phase retrieved from each satellite.
Abstract: Cloud optical thickness, also known as cloud optical depth, is an indicator that quantitatively shows the attenuation effect of solar radiance by clouds in the atmosphere and is calculated through statistical models or the radiative transfer model of each satellite using its observation data and ancillary data. Therefore, even for observations performed at the same location and time, the retrieved COT differ from satellite to satellite. Thus, this study aims to verify the retrieved COT of the GK-2A and HIMAWARI satellites, which are Korean and Japanese geostationary satellites, respectively, covering the Korean Peninsula. To verify the COT data, the method used by Qin et al. (2019) was used, where the COT was indirectly verified by calculating downward radiation using satellite-retrieved data such as COT, cloud phase, and cloud top pressure as parameters and comparing this with ground-observed radiation. In this study, a radiative transfer model, libRadtran 2.0.4, was used to calculate UV-A radiation and compare it with observed data in Daejeon as the true value for the cloud phase retrieved from each satellite. When comparing the COT from both satellites directly, the values from HIMAWARI tended to be larger than the data from GK-2A. A comparison of the UV-A radiation shows that the observed values are seemingly larger than the satellite results, indicating that both satellites may overestimate the cloud optical depth. Additionally, when both satellites were estimated to have the same cloud phase, HIMAWARI showed better parameters for RMSE and MAE, whereas GK-2A was better when GK-2A and HIMAWARI had different cloud phase estimates. By comparing the freezing level from the vertical profile and the cloud top height from each satellite, the actual cloud phase was estimated, which showed that GK-2A had better performance in estimating cloud phases.
TL;DR: In this paper , a volume photon mapping algorithm is proposed to depict the light field based on the volume of the photons in the environment, thus providing an unbiased luminance distribution, which enables an intuitive interpretation of the light propagation that helps designers to understand the basic light field in the space.
Abstract: This paper proposes a new method to depict the light field based on the volume photon mapping algorithm. In the context of the light field simulation, a participating medium serves to deposit the photons, but does not disturb their propagation. The photons are therefore neither scattered nor absorbed in order to preserve their energy and trajectory within the environment, thus providing an unbiased luminance distribution. A visualisation of the photon distribution enables an intuitive interpretation of the light propagation that helps designers to understand the basic light field in the space. In addition to visualisation, the magnitude of the simulated physical light field can be numerically evaluated from the volume photon map distribution using, for example, cubic and scalar illuminance. This can further inform the designer on the light density distribution in the space, since the latter directly correlates with the density of the photons, and therefore the scalar illuminance. The accuracy of the proposed method was ascertained by comparing it with the original RADIANCE. Furthermore, its advantage in visualisation was demonstrated using a complex case study involving strong indirect lighting, reinforced by a comparison of the simulation and measurement in the actual space. In addition, photon mapping was found to evaluate illuminance in multiple grid points much faster than RADIANCE Classic, notably due to the complex ambient lighting from specular reflections. The implementation of the specialised volume photon mapping software is now part of the RADIANCE software and is available as a lighting research tool for the community.
01 Jan 2013
TL;DR: In this article, a graph-based model is applied to the radiance data to capture the intricate structure of each material manifold, followed by the application of the commute time distance (CTD) transformation to separate the target manifold from the background.
Abstract: Identification of materials from calibrated radiance data collected by an airborne imaging spectrometer depends strongly on the atmospheric and illumination conditions at the time of collection. This thesis demonstrates a methodology for identifying material spectra using the assumption that each unique material class forms a lower-dimensional manifold (surface) in the higher-dimensional spectral radiance space and that all image spectra reside on, or near, these theoretic manifolds. Using a physical model, a manifold characteristic of the target material exposed to varying illumination and atmospheric conditions is formed. A graph-based model is then applied to the radiance data to capture the intricate structure of each material manifold, followed by the application of the commute time distance (CTD) transformation to separate the target manifold from the background. Detection algorithms are then applied in the CTD subspace. This nonlinear transformation is based on a random walk on a graph and is derived from an eigendecomposition of the pseudoinverse of the graph Laplacian matrix. This work provides a geometric interpretation of the CTD transformation, its algebraic properties, the atmospheric and illumination parameters varied in the physics-based model, and the influence the target manifold samples have on the orientation of the coordinate axes in the transformed space. This thesis concludes by demonstrating improved detection results in the CTD subspace as compared to detection in the original spectral radiance space.
TL;DR: In this article, the radiance and degree of polarization of atmospheric aerosols were measured with a portable photopolarimeter over the ocean around the Shikoku-Island of Japan in July, 1995.