China 
Africa 
Explore chapters and articles related to this topic
Introduction to Computer Vision and Basic Concepts of Image Formation
Published in Manas Kamal Bhuyan, Computer Vision and Image Processing, 2019
In spherical coordinate system, BRDF can be defined as follows: ρbd(θo,ϕo,θi,ϕi)=Lo(P,θo,ϕo)Li(P,θi,ϕi)cosθidω(sr−1) If we neglect subsurface scattering and transmission, then energy leaving a surface will be equal to the energy arriving the surface, i.e., symmetric in both the directions (Helmholtz reciprocity).
The Diverse Domain
Published in Aditi Majumder, M. Gopi, Introduction to Visual Computing, 2018
Speed: The speed of rendering a scene is always dependent on how much complexity is modeled and rendered. The most fundamental parameter that dictates speed is the number of primitives since it is inversely proportional to the speed of rendering. Complex phenomena like caustic effects or realistic illumination effects can make the rendering very slow. In popular terms, if the rendering can be achieved at a video rate, i.e. 30 frames per second (fps), it is called an interactive rendering. But, it should be kept in mind that the term interactive is also application dependent. For example, a game application may need to be rendered at 30 fps to be termed as interactive, but a sketch application can respond at 10 fps and the user may still feel that the system is responding appropriately to the sketch strokes. However, more often than not, rendering a frame for minutes or hours is termed as non-interactive. Most complex phenomena like subsurface scattering or cloth appearance modeling are usually associated with non-interactive rendering. Therefore, almost all the renderings in Figure 12.2 have taken multiple machines and many hours to render a single frame.
Applications of Spectral Imaging and Reproduction to Cultural Heritage
Published in Filippo Stanco, Sebastiano Battiato, Giovanni Gallo, Digital Imaging for Cultural Heritage Preservation, 2017
Filippo Stanco, Sebastiano Battiato, Giovanni Gallo
In computer graphics, various light reflection models have been proposed during the past decades. The first type of reflection models are based on theory of optics and physics. Torrance and Sparrow [130] proposed a model where rough surfaces are approximated through small, mirror-like, and randomly oriented facets. Blinn [131] adapted the Torrance-Sparrow in order to optimize the performance for computer graphics applications. Blinn substituted the Gaussian microfacet distributed function with an ellipsoid of revolution modeled function. Cook and Torrance [132] developed a more general reflection model which is able to approximate the color shift that occurs when the reflectance changes with the incident angle. He and Torrance [133] extended the Cook and Torrance model with polarization, directional Fresnel effects, and the subsurface scattering effect. Shirley et al. [134] presented an approach for polished materials, i.e., those materials having smooth surfaces and a significative subsurface scattering. Finally, the Beard-Maxwell model [135] approximated the material surface using a three-dimensional terrain of micro-facets. They also included shadowing, obscuration, and polarization effects.
An ANN-based surrogate model for wave propagation in uncertain media
Published in Waves in Random and Complex Media, 2023
Xi Cheng, Wei Shao, Kai Wang, Bing-Zhong Wang
The ground penetrating radar (GPR) [1–5] is an important remote sensing tool in many fields such as civil engineering [4], landmine detection [5], and environmental applications [6]. The substantial applications of GPR systems have created the need for a better understanding of subsurface-scattering mechanisms [2]. The numerical simulation is an effective approach for interpretation of wave propagation in GPRs. Some numerical methods have been employed for GPR system modeling [2]. Among these numerical methods, the finite-difference time-domain (FDTD) method [7] is one of the commonly used methods because it is easy to implement and it can also model dispersive and lossy media [8]. The numerical simulation of GPRs relies on a set of input parameters which can affect the electromagnetic pulses and then the survey of an object [2]. In practice, the exact values of the inputs are always unknown, leading to uncertainties in the output of the simulation [9]. Quantifying the uncertainty in simulation results is an indispensable part in GPR calculation when the acceptability of the output is considered [10].
Simultaneous estimation of projector and camera poses for multiple oneshot scan using pixel-wise correspondences estimated by U-Nets and GCN
Published in Computer Methods in Biomechanics and Biomedical Engineering: Imaging & Visualization, 2022
Ryo Furukawa, Michihiro Mikamo, Hiroshi Kawasaki, Ryusuke Sagawa, Shiro Oka, Takahiro Kotachi, Yuki Okamoto, Shinji Tanaka
For this study, a projector-camera system was constructed by inserting a fibre-shaped, micro-pattern projector into the instrument channel of a standard endoscope. We used a Fujifilm EG-590WR endoscope and a pattern projector with a diffractive optical element (DOE) to generate structured-light illumination. The pattern projector can be inserted into the endoscope’s instrument channel and patterns are projected from the projector to surfaces in front of the head of the endoscope (Figure 1). As shown in (Figure 1), we used a grid pattern that is robust against subsurface scattering Furukawa et al. (2016). All vertical edges are connected; horizontal edges have small gaps, representing code letters , and as shown as coloured codes in (Figure 1) (top right), where red dots mean that the right and the left edges of the grid point have the same height (code letter ), blue means the left side is higher (code letter ), and green means the right is higher (code ).
UAV-based mapping of nearshore bathymetry over broad areas
Published in Coastal Engineering Journal, 2020
Fumiya Tsukada, T. Shimozono, Y. Matsuba
The bathymetry estimation also requires a representative frequency of shoreward propagating waves. It must be determined from pixel intensity time-series used for wave celerity estimation for consistency. However, pixel intensity power spectrum is different from that of water surface fluctuations, since the light intensity changes result from complex optical processes on curved water surface such as reflection, subsurface scattering, and shadowing. Here the representative frequency was evaluated as a weighted average frequency by cross-spectrum magnitude between two signals along the cross-shore section. As the cross-spectrum reflects both amplitude and correlation of the two signals, the representative frequency is evaluated by assigning a larger weight to the large-amplitude coherent signals along the cross-shore section. This is more consistent with the wave celerity estimation than using the mean frequency of the power spectral density of signals. With the wave celerity variation and representative frequency along a cross-shore section, the cross-shore depth profile can be obtained by substituting these values into the dispersion relationship. Thus, a two-dimensional bathymetry was mapped by repeating this process at 2-m intervals in the longshore direction, assuming that swells propagate normally to the shoreline.