Christophe Renaud

Improving Light Position in a Growth Chamber through the Use of a Genetic Algorithm

Growth chambers are used by agronomists for various experiments on plants. Because light impinging on plants is one of the main parameters of their growth, inhomogeneity in light reception can provide large bias in the experiments. In this paper we present the first steps of a framework which aims at computing the best locations of light sources that could ensure an homogeneous lighting at some places in these chambers. For this purpose we extent the capabilities of a global illumination approach dedicated to growth chambers. We use Genetic Algorithms with simple source and material models in order to find good light sources locations. Our first results show that it is possible to find such interesting location and that they improve the lighting distribution on the experiment tables used in these chambers.

A no-reference computer-generated images quality metric and its application to denoising

A no-reference image quality metric detecting both blur and noise is proposed in this paper. The proposed metric is based on IFS2 entropy applied on computer-generated images and does not require any edge detection. Its value drops either when the test image becomes blurred or corrupted by random noise. It can be thought of as an indicator of the signal to noise ratio of the image. Experiments using synthetic, natural and computer-generated images are presented to demonstrate the effectiveness and robustness of this metric. The proposed measure has been too compared with full-reference quality measures (or faithfullness measures) like SSIM and gives satisfactory performance.

Massively parallel hemispherical projection for progressive radiosity

Abstract This paper describes a massively parallel implementation of the progressive radiosity algorithm. Our algorithm is based on an hemispherical projection approach, which provides and accurate form factor approximation. As the projection plane is mapped onto a processor mesh, we propose different techniques that decrease computation time by reducing processor inactivity as much as possible. This approach successfully handles large sets of form factor sampling elements.

A perceptual stopping condition for global illumination computations

The aim of realistic image synthesis is to produce high fidelity images that authentically represent real scenes. As these images are produced for human observers, we may exploit the fact that not everything is perceived when viewing scene with our eyes. Thus, it is clear that taking advantage of the limited capacity of the human visual system (HVS), can significantly contribute to optimize rendering software. Global illumination methods are used to simulate realistic lighting in 3D scenes. They generally provide a progressive convergence to high-quality solution. One of the problem of such algorithms is to determine a stopping condition, for deciding if calculations reached a satisfactory convergence allowing the process to terminate. In this paper, we propose and we discuss different solutions to this important problem. We show different techniques based on the Visual Difference Predictor (VDP) proposed by Daly [Daly 1993] to define a perceptual stopping condition for rendering computations. We use the VDP to measure the perceived differences between rendered images and to guide the Path Tracing rendering to satisfy a perceptual quality. Also, in a controlled experimental setting with real subjects, we validate our results.

Radiometric compensation for a low-cost immersive projection system

Catopsys is a low-cost projection system aiming at making mixed reality (virtual, augmented or diminished reality) affordable. It combines a videoprojector, a camera and a convex mirror and works in a non-specific room. This system displays an immersive environment by projecting an image onto the different parts of the room. However, the presence of an uncalibrated projector, heterogeneous materials and light inter-reflections influence the colors of the environment displayed in the room. Radiometric compensation of the projection process enables the system to reduce this problem. In this paper, we present our low-cost immersive projection system and propose a radiometric model and a compensation method which handle the projector response, surface materials and inter-reflections between surfaces. Our method works in two stages. First, the radiometric response of the projection process is evaluated. Then, this radiometric response is used to compensate the projection process in the desired environments.

Real-time animation of various flame shapes

Working on the computer reconstruction of the Gallo-Roman forum of Bavay, we try to improve the feeling of immersion in the virtual environment. One way to achieve this is to provide realistic and dynamic light sources. In this context, we need to model candles, oil lamps, torches or bonfires. We propose in this paper a model that can handle complex flames in real-time and manage interactivity. The fire is considered as a set of linear flames whose shapes are defined by the geometry of the combustible and the fuel distribution. Each individual flame is represented by a textured NURBS surface. Then, combining several real-time effects such as glow and true transparency, we are able to make the NURBS surfaces merge in a convincing way, and to give the impression of a real fire.

A progressive mesh method for physical simulations using lattice Boltzmann method on single-node multi-gpu architectures

In this paper, a new progressive mesh algorithm is introduced in order to perform fast physical simulations by the use of a lattice Boltzmann method (LBM) on a single-node multi-GPU architecture. This algorithm is able to mesh automatically the simulation domain according to the propagation of fluids. This method can also be useful in order to perform various types of simulations on complex geometries. The use of this algorithm combined with the massive parallelism of GPUs allows to obtain very good performance in comparison with the static mesh method used in literature. Several simulations are shown in order to evaluate the algorithm.

Detecting Visual Convergence for Stochastic Global Illumination

Photorealistic rendering, based on unbiased stochastic global illumination, is now within reach of any computer artist by using commercially or freely available softwares. One of the drawbacks of these softwares is that they do not provide any tool for detecting when convergence is reached, relying entirely on the user for deciding when stopping the computations. In this paper we detail two methods that aim at finding perceptual convergence thresholds for solving this problem. The first one uses the VDP image quality measurement for providing a global threshold. The second one uses SVM classifiers which are trained and used on small subparts of images and allow to take into account the heterogeneity of convergence through the image area. These two approaches are validated by using experimentations with human subjects.

Light Source Storage and Interpolation for Global Illumination: A Neural Solution

Photorealistic lighting requires to use accurate rendering algorithms and realistic models of light sources. Photometric solids are provided by lighting design- ers in order to characterize the luminance distribution from real light sources. When accuracy is required the amount of discretized luminance directions and the num- ber of photometric solids that have to be used increase considerably the memory requirements and reduce the algorithms efficiency. In this paper we propose to ap- proximate any photometric solid by the way of artificial neural networks. The data describing the luminances distribution are used to train a dedicated neural network during a learning step. Then this much more compact representation of the data can be used both for reducing the memory requirements and the computation costs when searching for any luminance information.

Real-time rendering and animation of plentiful flames

Rendering and animating flames in real time is a great challenge because of the complexity of the combustion process. While few models succeeded in simulating a single fire in real time, none tried to handle a large number of flames. In this paper we propose a set of techniques which aim at handling numerous flames like those of candles, torches or campfires in real time. We manage different levels of accuracy in the simulation, which is based on a fast fluid dynamics solver. We also consider many optimizations to reduce the rendering cost of a single flame. As a result, our approach is able to animate and render dozens of realistic flames in real-time, each one reacting independently to forces introduced at the scene level.