Search

Data-driven BRDF models using real material measurements have become increasingly prevalent due to the development of novel gonioreflectometers, but efficient use of these models in many graphical applications remains challenging due to the few functionalities the raw data could provide. To ameliorate this issue, we propose to analyze BRDFs using directional statistics for better handling and exploring measured materials, especially isotropic materials, with efficient computation and compact storage. We conduct a thorough statistical analysis on both analytical BRDF models and measured materials from the MERL database. We show that different aspects of visual appearance can be characterized by different spherical moments, from which several descriptive measures can be derived to further facilitate their usage. We demonstrate how these measures are best leveraged in some graphical applications including gamut mapping using a new BRDF similarity measure, BRDF or SVBRDF reconstruction based on material clustering, and importance sampling for measured materials based on fast extracted GGX distributions. We finally show the potential of our approach in the categorization of surface reflectance types which is common for traditional photon mapping.

In this paper, we present an efficient framework for editing the appearance of measured materials. First, we separate the material data into diffuse and specular parts, which is important for measured BRDFs since editing without distinguishing the two parts makes the results less predictive. Additionally, the albedo of each part is calculated for later color editing. Next, we use a novel method to cluster the materials by their gloss levels and select the representative material of every level. With this at hand, the roughness, i.e., shape of the specular highlights can be adjusted among different levels. Finally, we reconstruct the edited materials which can be directly used in both online and offline rendering. We test the framework on the MERL dataset to validate the effectiveness of our method.

As taxi service is supervised by certain electronic equipment (e.g., global positioning system (GPS) equipment) and network technique (e.g., cab reservation through Uber in USA or DIDI in China), taxi business is a typical electronic commerce mode. For a long time, taxi service is facing a typical challenge, that is, passengers may be detoured and overcharged by some unethical taxi drivers, especially when traveling in unfamiliar cities. As a result, it is important to detect taxi drivers’ misbehavior through taxi’s GPS big data analysis in a real-time manner for enhancing the quality of taxi services. In view of this challenge, an online anomalous trajectory detection method, named OnATrade (pronounced “on a trade,” which means activities in a taxi trade on the fly), is investigated in this paper for improving taxi service using GPS big data. The method mainly consists of two steps: route recommendation and online detection. In the first step, route candidates are generated by using a route recommendation algorithm. In the second step, an online anomalous trajectory detection approach is presented to find taxis that have driving anomalies. Experiments evaluate the validity of our method on large-scale, real-world taxi GPS trajectories. Finally, several value-added applications benefiting from big data analysis over taxi’s GPS data sets are discussed for potential commercial applications.

In this paper, we propose a depth restoration method using a novel superpixel technique. Guided by a normal map reconstructed from the raw depth data, this technique over-segments RGB-D images into many small regions where their depth is assumed to be smooth. As the raw depth data is incomplete, we further introduce a depth confidence map to identify the regions which are more reliable. With the produced superpixels, we can restore the incomplete depth map using a per-superpixel linear regression. A multiresolution su-perpixel strategy is employed when some superpixels do not contain enough valid data. Experiments show that the proposed depth restoration method can effectively fill the wide gaps along depth discontinuities without blurring the object boundaries and the depth discontinuities.

This paper introduces a robust method for removing objectionable reflection interference in photographs captured through a piece of transparent medium. We exploit the fact that a group of image patches extracted from multiple correlated images with similar transmission lie in a very low-dimensional subspace, leading to a low-rank matrix after patch assembly. This allows us to formulate reflection separation as a per-patch sparse and low-rank decomposition problem which can be well solved by the ALM-ADM strategy. To eliminate the influence of unwanted reflection in patch searching and ensure that the extracted patches has a high similarity regarding their transmission layers, we introduce a new patch similarity metric based on both image intensities and gradients. This improves the performance of reflection separation. In addition, since our method does not require image reconstruction from gradient, color-shifting artifacts can be significantly ameliorated and more scene details can be preserved. Experimental results on both synthetic images and various real-world examples demonstrate that the proposed method achieves high quality reflection separation and performs favorably against many existing techniques.

Most traditional Chinese paintings are painted on hand-made paper that easily suffers from severe spectral changes caused by prolonged light exposure, resulting in color distortion and low contrast. To recover the original appearance of a Chinese painting, especially its background color, an automatic background adjustment framework is proposed. This framework is based on the insightful observation that the fading model of a painting image is analogue to the common hazy image formation model when the painting image is transformed into the K-M (Kubelka-Munk) space. We demonstrate that this fading model is quite useful in extracting pigment lines from any painting image. These pigment lines represent clusters of distinct color pigments used in a Chinese painting, which is the key to density map estimation and background restoration. Experimental results prove that our approach is able to restore a variety of deteriorated Chinese paintings without any user intervention or training.

This paper presents an efficient approach to render refracting transmissive objects with multi-scale surface roughness, under distant illumination. To correctly capture both fine-scale surface details and large-scale appearances, and enable real-time processing at various viewing resolutions, we first divide the surface roughness into three levels, namely, micro-scale, meso-scale and macro-scale. Each scale of roughness is modeled and evaluated using different strategies, and the overall roughness is approximated by their spherical convolution. Then, this representation is incorporated into a microfacet-based BTDF model, and multi-scale rough refractions are simulated on both front and back sides of an object as light enters and exits the object. In particular, non-linear filtering methods are applied to both macro-scale geometries and meso-scale bumps to reduce aliasing when viewed across a range of distances. Finally, experimental results illustrate that our approach produces resolution-dependent refraction effects that match super-sampled ground truth, while achieving a speed up of several orders of magnitude with hardware acceleration.

BRDF importance sampling, which generates sampled directions in a pattern that closely matches the BRDF, is a powerful tool for variance reduction in Monte Carlo rendering. Conventionally, it is challenging to design proper sampling patterns for measured BRDFs since no analytical expression is available directly. Although tabulation based sampling strategy provides a feasible solution to this problem, it usually requires a high memory consumption for additionally storing the importance functions. In this poster, we show that the second order spherical moment of a BRDF can be leveraged in deriving a robust sampling function for the BRDF. This sampling function has an analytical form of a GGX distribution which resembles the shape of the raw BRDF measure. Besides efficient computation and compact storage brought by the GGX distribution, another benefit is its possibility to perform sampling according to the distribution of visible normals. This can further reduce the variance especially for diffuse-like materials and improve the convergence rate of the Monte Carlo numerical integration.

Detailed modeling of appearance affected by dust coverage is challenging because of the numerous physical and material factors on which it depends. The two principal roadblocks are the complexity of simulating dust particle deposition and the difficulty of accurately depicting the reflectance. In this paper, we present a practical framework for simulating and rendering the appearance of objects covered by the layered dust of spatially varying thickness. First, a fast evaluation approach is proposed to compute the dust's thickness distribution based on a surface's inclination and stickiness properties, as well as surface exposure and local stability. Then, to ensure high fidelity of light scattering and real-time performance, we render the scene based on a revised Kubelka---Munk model implemented on the GPU, including the effects of external and internal surface reflection. Finally, experimental results reveal that our framework can produce visually convincing dusty objects of arbitrary complexity in real time.

Web service recommendation plays an important role in building reliable service-oriented systems for both the service providers and the active users. However, with the proliferation of web services on the World Wide Web, traditional service recommendation is hard to accurately provide customized services to active users. In this paper, we propose a novel web service recommender model using collaborative filtering to improve the prediction of Quality-of-Services. Benefiting from the accuracy of hybrid recommenders, we extend the idea of optimized predicting order and design the Graph to describe the neighborhood. Furthermore, a new algorithm using adjusted topological sorting for Graph is proposed to generate the optimized order while predicting. Finally, we conduct extensive experiments to evaluate our proposed model, in which a real data set with 1.5 million invocation information is taken as input. The experiment results show that our model achieves higher prediction accuracy than other models.

We present a novel approach for real-time rendering of transparent objects with multi-scale rough refraction under all-frequency environmental lighting. In our rendering system, the normal distribution function (NDF) of the surface bumps is expressed via mixtures of von Mises-Fisher (vMF) distributions, while the corresponding perturbation of refracted rays (RDF) can be approximated as a spherical warp of NDF, taking into account correction factors. Under the assumption of two-sided refraction, we first generate the vMF lobes of the front surface refraction. Based on these lobes, we then estimate the RDF of back surface as a warped convolution of macroscopic and microscopic NDFs within the footprint of each lobe. Moreover, summed-area tables and dual-paraboloid maps are adopted to efficiently represent the environmental lighting in order to support fully dynamic lighting conditions. We finally get a coherent rendering at the different scales that reproduces high-quality meso- and micro-geometric effects under real-time performance.

The illumination of rough refraction is challenging due to the complex integrations of scattered lights from multiple directions. Prefiltered environment map is able to eliminate the comprehensive integrations, but the precomputation stage protects it from being adapted into dynamically varying environmental lighting. We propose a real-time method for rendering rough refraction, supporting dynamically changing lighting, geometries and materials. Environment maps are efficiently compressed by summed area table (SAT), whose construction only costs several milliseconds. Meanwhile, rough refraction effects are approximated by our lobe merging method. Solid results of our approach are demonstrated by comparing with ground truth results generated with a path tracing engine.

Recommending suitable routes to taxi drivers for picking up passengers is helpful to raise their incomes and reduce the gasoline consumption. In this paper, a pick-up tree based route recommender system is proposed to minimize the traveling distance without carrying passengers for a given taxis set. Firstly, we apply clustering approach to the GPS trajectory data of a large number of taxis that indicates state variance from “free” to “occupied”, and take the centroids as potential pick-up points. Secondly, we propose a heuristic based on skyline computation to construct a pick-up tree in which current position is its root node that connects all centroids. Then, we present a probability model to estimate gasoline consumption of every route. By adopting the estimated gasoline consumption as the weight of every route, the weighted Round-Robin recommendation method for the set of taxis is proposed. Our experimental results on real-world taxi trajectories data set have shown that the proposed recommendation method effectively reduce the driving distance before carrying passengers, especially when the number of cabs becomes large. Meanwhile, the time-cost of our method is also lower than the existing methods.

Complex natural lighting provides better realism compared to traditional artificial lights, which has been increasing in real-time graphic applications, such as video games and virtual reality. However, rendering with complex natural illumination proves quite costly owing to a spherical space integral. This paper addresses the problem of real-time rendering scenes under complex environment lighting that captures soft shadows and glossy reflections. The environment lighting is separated into low-frequency lighting component and highlight component based on a rejection-based sampling method. Scenes lit under low-frequency lighting are approximated via spherical harmonic lighting, while highlight illumination effects such as soft shadows and glossy reflections are simulated by a small set of sampled virtual point lights (VPL). Experimental results demonstrate that our method can be run in real-time under complex natural lighting without precomputation.

Dynamic local area lights provide better realism compared with traditional artificial lights, which have been increasing in real-time graphic applications. This paper addresses the problem of real-time direct lighting under dynamic local area lighting by subdividing the original complex area lights into several virtual area lights (VALs) which have uniform radiosity. Then these VALs are used to illuminate the scenes in a closed form solution. Experimental results demonstrate that our method can be run in real-time without precomputation.

In a multi-service distibuted virtual environment, objects can provide more services and become dynamic with greater effects in load balancing process, which is not considered in the previous work. Focused on the impact of dyanmic objects and their relationships with avatars to form workload-intensive groups when doing load balance, this paper proposes a group-based load balancing approach. Evaluation results show our approach can keep the system with dynamic objects stable regardless of the activities and increase of avatars in the system.