The gaze information channel paradigm models fixation sequences as a first-order Markov chain and quantifies gaze behaviour through Shannon entropy and mutual information (MI), where I(X;Y) measures the reduction in uncertainty about the next fixation state given the current one. This paper extends the framework by introducing two new channels: the saccade amplitude channel, which discretises saccade angular distance into three categories (short, medium, long) with a four-category variant also analysed, and the pupil diameter channel, which discretises fixation-period pupil size into three categories. Both are applied to 10 observers viewing 12 Van Gogh paintings. The amplitude channel shows that observer-driven variation exceeds stimulus-driven variation. The pupil channel yields the highest I(X;Y) among the two new channels (0.489±0.209 bits per participant), consistent with the slow dynamics of pupil responses. Goodness-of-fit tests confirm significantly non-random sequential structure in both channels (p<0.01) for all pooled matrices. A simultaneous cross-channel association analysis across all five channels finds that 19 of 20 pairwise Spearman correlations are non-significant; the single nominally significant result (pupil-duration, ρ=+0.697, p=0.025) does not survive Bonferroni correction and is not robust to outlier removal. Two theoretical observations are presented: an upper bound on conditional entropy in terms of transition persistence (Proposition 1), and a refinement monotonicity result showing that finer discretisation cannot decrease channel MI (Remark 2). An exploratory comparison with five computational aesthetics measures finds a nominally significant negative correlation between pupil I(X;Y) and Bense's palette redundancy (ρ=-0.692, p=0.013, uncorrected), suggesting that diverse colour palettes are associated with stronger sequential pupil dynamics; permutation entropy and statistical complexity show no association with any channel.

Multiple importance sampling is an efficient Monte Carlo method to approximate integrals. It is able to simulate samples from several proposal distributions and then weight these samples accordingly. Different weighting strategies have been proposed in the literature, some of which largely reducing the variance of the Monte Carlo estimators. However, obtaining good weighting mechanisms is often challenging due to intractable integrals and non-convex optimization problems. In this work, we propose two new weighting mechanisms that are both efficient to design and obtain provable high performance. We build upon our recent work, where quasi optimal weights are obtained by solving a linear equation. Our new methods lead to more efficient computations and more robust estimations through a fast convex minimization of the variance of the estimator. The contribution of this paper is also to provide a better understanding of previous works. We validate the new methods in three experiments, showing their excellent performance.

Virtual reality (VR) rehabilitation has been proven to be a very promising method to increase the focus and attention of patients by immersing them in a virtual world, and through that, improve the effectiveness of the rehabilitation. One of the biggest challenges in designing VR Rehabilitation exercises is in choosing feedback strategies that guide the patient and give the appropriate success/failure indicators, without breaking their sense of immersion. A new strategy for feedback is proposed, using non-photorealistic rendering (NPR) to highlight important parts of the exercise the patient needs to focus on and fade out parts of the scene that are not relevant. This strategy is implemented into an authoring tool that allows rehabilitators specifying feedback strategies while creating exercise profiles. The NPR feedback can be configured in many ways, using different NPR schemes for different layers of the exercise environment such as the background environment, the non-interactive exercise objects, and the interactive exercise objects. The main features of the system including the support for universal render pipeline, camera stacking, and stereoscopic rendering are evaluated in a testing scenario. Performance tests regarding memory usage and supported frames per second are also considered. In addition, a group of rehabilitators evaluated the system usability. The proposed system meets all the requirements to apply NPR effect in VR scenarios and solves all the limitations with regard to technical function and image quality. In addition, the system performance has been shown to meet the targets for low-cost hardware. Regarding authoring tool usability rehabilitators agree that is easy to use and a valuable tool for rehabilitation scenarios. NPR schemes can be integrated into VR rehabilitation scenarios achieving the same image quality as non-VR visualizations with only a small impact on the frame rate. NPR schemes are a good visual feedback alternative.

Eye tracking has become an increasingly important technology in many fields of research, such as marketing, human computer interaction, psychology, and also in human cognition. Understanding the human eye movements, while viewing specific scenarios, can be of great support for improving visual stimuli. However, the challenging problem with this kind of spatio-temporal data is to find quantitative links between eye movements and human cognition. This paper introduces the information channel based on saccade direction. The gaze transition between different saccade directions is modeled as a discrete information channel, which we call saccade direction information channel . The channel is applied to an eye-tracking dataset on Van Gogh’s paintings observation. In our results, horizontal saccades are more frequent than vertical saccades, and the information conveyed in horizontal/vertical displacements, measured as the mutual information of the channel, is higher than in diagonal displacements. By comparing the results to our previous spatial gaze channel between Areas of Interest (AOIs) we constate that the spatial channel discriminates better between the observed images, while the direction channel discriminates better between observers.

Multi-exposure image fusion has emerged as an increasingly important and interesting research topic in information fusion. It aims at producing an image with high quality by fusing a set of differently exposed images. In this article, we present a pixel-level method for multi-exposure image fusion based on an information-theoretic approach. In our scheme, an information channel between two source images is used to compute the Renyi entropy associated with each pixel in one image with respect to the other image and hence to produce the weight maps for the source images. Since direct weight-averaging of the source images introduce unpleasing artifacts, we employ Laplacian multi-scale fusion. Based on this pyramid scheme, images at every scale are fused by weight maps, and a final fused image is inversely reconstructed. Multi-exposure image fusion with the proposed method is easy to construct and implement and can deliver, in less than a second for a set of three input images of size 512x340, competitive and compelling results versus state-of-art methods through visual comparison and objective evaluation.

When viewing the surrounding environment, the human eyes perform different eye movements that reflect human's interest and attention. A large amount of research efforts have focused on Areas of Interest (AOIs)-based transition analysis of eye movement. However, to the best of our knowledge, there is little work on fixation duration from eye tracking data. In this paper, we investigate entropy based eye movement transition analysis of fixation duration. Specifically, we divide the fixation duration into three categories - express fixations (called short duration, SD), cognitive fixations (called medium duration, MD), and very long fixations (called long duration, LD), and model the eye tracking sequences of fixation duration as a Markov chain, which is further considered from the perspective of a discrete information channel. Subsequently, we exploit the information channel paradigm by computing the entropy measures from the fixation duration information channel to explore the visual behaviour. The proposed method, that allows a straight-forward clustering of the results, is demonstrated on two eye tracking data sets, Van Gogh paintings and scientific posters, and shows that the differences between observers are more pronounced that the differences between the observed images. The findings from this work might also provide future insights to quantitative assess human visual cognitive process.

Rendering inhomogeneous participating media requires a lot of volume samples since the extinction coefficient needs to be integrated along light paths. Ray marching makes small steps, which is time consuming and leads to biased algorithms. Woodcock-like approaches use analytic sampling and a random rejection scheme guaranteeing that the expectations will be the same as in the original model. These models and the application of control variates for the extinction have been successful to compute transmittance and single scattering but were not fully exploited in multiple scattering simulation. Our paper attacks the multiple scattering problem in heterogeneous media and modifies the light-medium interaction model to allow the use of simple analytic formulae while preserving the correct expected values. The model transformation reduces the variance of the estimates with the help of Rao-Blackwellization and control variates applied both for the extinction coefficient and the incident radiance. Based on the transformed model, efficient Monte Carlo rendering algorithms are obtained.

Blind players have many difficulties to access video games since most of them rely on impressive graphics and immersive visual experiences. To overcome this limitation, we propose a device designed for visually impaired people to interact with virtual scenes of video games. The device has been designed considering usability, economic cost, and adaptability as main features. To ensure usability, we considered the white cane paradigm since this is the most used device by the blind community. Our device supports left to right movements and collision detection as well as actions to manipulate scene objects such as drag and drop. To enhance realism, it also integrates a library with sounds of different materials to reproduce object collision. To reduce the economic cost, we used Arduino as the basis of our development. Finally, to ensure adaptability, we created an application programming interface that supports the connection with different games engines and different scenarios. To test the acceptance of the device 12 blind participants were considered (6 males and 6 females). In addition, we created three mini-games in Unity3D that require navigation and walking as principal actions. After playing, participants filled a questionnaire related to usability and suitability to interact with games, among others. They scored well in all features without distinction among player gender and being blind from birth. The relationship between device responsiveness and user interaction has been considered satisfactory. Despite our small test sample, our main goal has been accomplished, the proposed device prototype seems to be useful to visually impaired people.

We propose a method for multi-exposure image fusion based on information-theoretic channel. In the fusion scheme, conditional entropy, as an information measurement of each pixel in one image to the other image, is calculated through an information channel built between two source images, and then weight maps of the source images are generated. Considering the noise caused by blending source images with weight maps directly, we exploit Laplacian pyramid decomposition to avoid unpleasing artifacts. Based on this pyramid scheme, images at every scale are fused by weight maps, and a final fused image is inversely reconstructed. The proposed method is easy to implement and delivers results which are competitive with state-of-art methods.

Cardiopulmonary resuscitation (CPR) is a first-aid key survival technique used to stimulate breathing and keep blood flowing to the heart. Its effective administration can significantly increase the survival chances of cardiac arrest victims. We propose 30 : 2, a videogame designed to introduce the main steps of the CPR protocol. It is not intended for certification and training purpose. Driven by the 2010 European Resuscitation Council guidelines we have designed a game composed of eight mini games corresponding to the main steps of the protocol. The player acts as a helper and has to solve a different challenge. We present a detailed description of the game creation process presenting the requirements, the design decisions, and the implementation details. In addition, we present some first impressions of our testing users (25 children, five of each age from 8 to 12 years old and 12 males and 13 females). We evaluated clarity of instructions and three settings of the game: the aesthetics of scenarios, the playability, and the enjoyability of each mini game. All games were well punctuated, and there are no significantly differences between their sex. The proposed game can be a suitable tool to disseminate and promote CPR knowledge.

In this paper, we present an information‐theoretic framework to compute the shape similarity between 3D polygonal models. Given a 3D model, an information channel between a sphere of viewpoints around the model and its polygonal mesh is defined to compute the specific information associated with each viewpoint. The obtained information sphere can be seen as a shape descriptor of the model. Then, given two models, their similarity is obtained by performing a registration process between the corresponding information spheres. The distance between the information histograms is also defined as a coarse measure of similarity, as well as the scalar value given by the mutual information of the channel. The performance of all these measures is tested using the Princeton Shape Benchmark database. Copyright © 2013 John Wiley & Sons, Ltd.

The classification of surface reflectance functions as diffuse, specular, and glossy has been introduced by Heckbert more than two decades ago. Many rendering algorithms are dependent on such a classification, as different kinds of light transport will be handled by specialized methods, for example caustics require specular bounce or refraction. Due to the increasing wealth of surface reflectance models including those based on measured data, it has not been possible to keep such a characterization simple. Each surface reflectance model is mostly handled separately, or alternatively, the rendering algorithm restricts itself to the use of some subset of reflectance models. We suggest a characterization for arbitrary surface reflectance representation by standard statistical tools, namely normalized variance known as Squared-Coefficient-of-Variation (SCV). We show by videos that there is even a weak perceptual correspondence with the proposed reflectance characterization, when we use monochromatic surface reflectance and the images are normalized so they have the unit albedo.

In this paper we present a method for radiosity computation in dynamic scenes. The algorithm is intended for animations in which the motion of the objects is known in advance. Radiosity is computed using a Monte Carlo approach. Instead of computing each frame separately, we propose to compute the lighting simulation of a sequence of frames in a unique process. This is achieved by the merging of the whole sequence of frames into a single scene, so each moving object is replicated as many times as frames.We present results which show the performance of the proposed method. This is specially interesting for sequences of a significant number of frames. We also present an analysis of the algorithm complexity. An important feature of the algorithm is that the accuracy of the image in each frame is the same as the one we would obtain by means of computing each frame separately.

Finding an optimal discretization of a scene is an important but difficult problem in radiosity. The efficiency of hierarchical radiosity for instance, depends entirely on the subdivision criterion and strategy that is used. We study the problem of adaptive scene discretization from the point of view of information theory. In previous work, we have introduced the concept of mutual information, which represents the information transfer or correlation in a scene, as a complexity measure and presented some intuitive arguments and preliminary results concerning the relation between mutual information and scene discretization. In this paper, we present a more general treatment supporting and extending our previous findings to the level that the development of practical information theory-based tools for optimal scene discretization becomes feasible.

Viewpoint selection is an emerging area in computer graphics with applications in fields such as scene exploration, image-based modeling, and volume visualization. In particular, best view selection algorithms are used to obtain the minimum number of views (or images) in order to understand or model an object or scene better. In this article, we present a unified framework for viewpoint selection and mesh saliency based on the definition of an information channel between a set of viewpoints (input) and the set of polygons of an object (output). The mutual information of this channel is shown to be a powerful tool to deal with viewpoint selection, viewpoint stability, object exploration and viewpoint-based saliency. In addition, viewpoint mutual information is extended using saliency as an importance factor, showing how perceptual criteria can be incorporated to our method. Although we use a sphere of viewpoints around an object, our framework is also valid for any set of viewpoints in a closed scene. A number of experiments demonstrate the robustness of our approach and the good behavior of the proposed measures.

This paper presents a real-time global illumination method for quasi-static scenes illuminated by arbitrary, dynamic light sources. In order to maintain real-time frame-rates, the light transport problem is decomposed to low-frequency and high-frequency parts. The part that has low-frequency characteristics both in the temporal and in the spatial domains is pre-computed. Due to the low-frequency properties this solution can be compressed by clustered PCA without significant reduction of the accuracy. The spatial high-frequency but temporal low-frequency parts are pre-computed on small-scale using a modified ambient occlusion/obscurances method. The temporally high-frequency components are obtained on the fly. This way we can combine the advantages of pre-computation aided methods and real-time rendering.

We propose a new viewpoint-based simplification method for polygonal meshes, driven by several f-divergences such as Kullback–Leibler, Hellinger and Chi-Square. These distances are a measure of discrimination between probability distributions. The Kullback–Leibler distance between the projected and the actual area distributions of the polygons in the scene already has been used as a measure of viewpoint quality. In this paper, we use the variation in those viewpoint distances to determine the error introduced by an edge collapse. We apply the best half-edge collapse as a decimation criterion. The approximations produced by our method are close to the original model in terms of both visual and geometric criteria. Unlike many pure visibility-driven methods, our new approach does not completely remove hidden interiors in order to increase the visual quality of the simplified models. This makes our approach more suitable for applications which require exact geometry tolerance but also require high visual quality.