Abstract Sketch‐based image retrieval (SBIR) is an emerging task in computer vision. Research interests have arisen in solving this problem under the realistic and challenging setting of zero‐shot learning. Given a sketch as a query, the search goal is to retrieve the corresponding photographs in a zero‐shot scenario. In this paper, we divide the aforementioned challenging work into three tasks and propose a sharing model framework that addresses these problems. First, the weights of the proposed sharing model effectively reduced the modality gap between sketches and photographs. Second, semantic information was used to handle different label spaces during the training and testing stages. The sketch and photograph domains share semantic information. Finally, a memory mechanism is used to reduce the intrinsic variety in sketches, even if they all belong to the same class. Sketches and photographs dominate the embeddings in turn. Because sketches are not limited by language, our ultimate goal is to find a method to replace text searches. We also designed a demonstration program to demonstrate the use of the proposed method in real‐world applications. Our results indicate that the proposed method exhibits considerably higher zero‐shot SBIR performance than do other state‐of‐the‐art methods on the challenging Sketchy, TU‐Berlin, and QuickDraw datasets.

Abstract Photos contain well‐structured and plentiful visual information. Edges are active and expressive stimuli for human visual perception. However, it is hard to separate structure from details because edge strength and object scale are entirely different concepts. This paper proposes a structure‐aware bilateral texture algorithm to remove texture patterns and preserve structures. Our proposed method is simple and fast, as well as effective in removing textures. Instead of patch shift, smaller patches represent pixels located at structure edges, and original patches represent the texture regions. Specifically, this paper also improves joint bilateral filter to preserve small structures. Moreover, a windowed inherent variation is adapted to distinguish textures and structures for detecting structure edges. Finally, the proposed method produces excellent experimental results. These results are compared to some results of previous studies. Besides, structure‐preserving filtering is a critical operation in many image processing applications. Our proposed filter is also demonstrated in many attractive applications, such as seam carving, detail enhancement, artistic rendering, etc.

Ray marching is an important technique to generate volumetric lighting effects. However, it is very expensive for each pixel on the screen, especially in dynamic scenes. The authors propose an adaptive approach to reduce samples for volumetric shadows in the time domain. In dynamic scenes, shadow volumes of moving objects are created and are rasterised to decide pixels that cannot be reused. The authors use a stencil buffer to maintain the information of screen pixels by recomputing or just using the previous information. Experimental results show that the proposed approach is simple to implement. Moreover, compared to the previous method, the proposed approach achieves a specific speedup and maintains similar visual quality.

With the increasing demands for photo-realistic image synthesis in real time, we propose a sparse multilinear model, which is named K-Clustered Tensor Approximation (K-CTA), to efficiently analyze and approximate large-scale multidimensional visual datasets, so that both storage space and rendering time are substantially reduced. K-CTA not only extends previous work on Clustered Tensor Approximation (CTA) to exploit inter-cluster coherence, but also allows a compact and sparse representation for high-dimensional datasets with just a few low-order factors and reduced multidimensional cluster core tensors. Thus, K-CTA can be regarded as a sparse extension of CTA and a multilinear generalization of sparse representation. Experimental results demonstrate that K-CTA can accurately approximate spatially varying visual datasets, such as bidirectional texture functions, view-dependent occlusion texture functions, and biscale radiance transfer functions for efficient rendering in real-time applications.

The motion of comic characters includes different types of movements, such as walking or running. In a comic, a movement may be described by a series of non-continuous poses in a sequence of contiguous frames. Each pose exists in a frame. We synthesize an animation according to still comic frames. In this paper, we propose a model to analyze time series of a character's motion using the non-parametric Bayesian approach. Then we can automatically generate a sequence of motions by using the estimated time series. Experimental results show that the built time series model best matches the given frames. Furthermore, unnatural distortions of the results are minimized. Copyright © 2011 John Wiley & Sons, Ltd.

This paper presents a novel parametric representation for bidirectional texture functions. Our method mainly relies on two original techniques, namely, multivariate spherical radial basis functions (SRBFs) and optimized parameterization. First, since the surface appearance of a real-world object is frequently a mixed effect of different physical factors, the proposed sum-of-products model based on multivariate SRBFs especially provides an intrinsic and efficient representation for heterogenous materials. Second, optimized parameterization particularly aims at overcoming the major disadvantage of traditional fixed parameterization. By using a parametric model to account for variable transformations, the parameterization process can be tightly integrated with multivariate SRBFs into a unified framework. Finally, a hierarchical fitting algorithm for bidirectional texture functions is developed to exploit spatial coherence and reduce computational cost. Our experimental results further reveal that the proposed representation can easily achieve high-quality approximation and real-time rendering performance.

Image-based rendering has been successfully used to display 3-D objects for many applications. A well-known example is the object movie, which is an image-based 3-D object composed of a collection of 2-D images taken from many different viewpoints of a 3-D object. In order to integrate image-based 3-D objects into a chosen scene (e.g., a panorama), one has to meet a hard challenge--to efficiently and effectively remove the background from the foreground object. This problem is referred to as multiview images (MVIs) segmentation. Another task requires MVI segmentation is image-based 3-D reconstruction using multiview images. In this paper, we propose a new method for segmenting MVI, which integrates some useful algorithms, including the well-known graph-cut image segmentation and volumetric graph-cut. The main idea is to incorporate the shape prior into the image segmentation process. The shape prior introduced into every image of the MVI is extracted from the 3-D model reconstructed by using the volumetric graph cuts algorithm. Here, the constraint obtained from the discrete medial axis is adopted to improve the reconstruction algorithm. The proposed MVI segmentation process requires only a small amount of user intervention, which is to select a subset of acceptable segmentations of the MVI after the initial segmentation process. According to our experiments, the proposed method can provide not only good MVI segmentation, but also provide acceptable 3-D reconstructed models for certain less-demanding applications.

Research on creating a large, high-resolution, low-cost display system has become increasingly important due to the growing desire in many fields for bigger and better displays. The goal of this research is to build a seamless large-scale display system by tiling multiple projectors with the help of a pan-tilt-zoom camera. In order to achieve this goal, we took two tasks into consideration: geometric calibration and photometric calibration. Compared to the previous work, our method for geometric calibration is more accurate, thanks to the much higher resolution images acquired by combining several zoom-in images and doing lens correction for the camera at first. For photometric calibration, we achieve color uniformity among projectors by utilizing the same PTZ camera, which is calibrated once with a colorimeter in a factory. Furthermore, we adopted the technique of producing a high dynamic range image from several images of different exposures to increase the measurement accuracy of the camera. In our experiments, the average error of photometric measurement achieved by our method is less than the mean perceptibility tolerance. Our method has great potential for many applications that require large and high-resolution displays

Watershed segmentation has recently become a popular tool for image segmentation. There are two approaches to implementing watershed segmentation: immersion approach and toboggan simulation. Conceptually, the immersion approach can be viewed as an approach that starts from low altitude to high altitude and the toboggan approach as an approach that starts from high altitude to low altitude. The former seemed to be more popular recently (e.g., Vincent and Soille), but the latter had its own supporters (e.g., Mortensen and Barrett). It was not clear whether the two approaches could lead to exactly the same segmentation result and which approach was more efficient. In this paper, we present two "order-invariant" algorithms for watershed segmentation, one based on the immersion approach and the other on the toboggan approach. By introducing a special RIDGE label to achieve the property of order-invariance, we find that the two conceptually opposite approaches can indeed obtain the same segmentation result. When running on a Pentium-III PC, both of our algorithms require only less than 1/30 s for a 256 x 256 image and 1/5 s for a 512 x 512 image, on average. What is more surprising is that the toboggan algorithm, which is less well known in the computer vision community, turns out to run faster than the immersion algorithm for almost all the test images we have used, especially when the image is large, say, 512 x 512 or larger. This paper also gives some explanation as to why the toboggan algorithm can be more efficient in most cases.

This paper introduces a new data representation and compression technique for precomputed radiance transfer (PRT). The light transfer functions and light sources are modeled with spherical radial basis functions (SRBFs). A SRBF is a rotation-invariant function that depends on the geodesic distance between two points on the unit sphere. Rotating functions in SRBF representation is as straightforward as rotating the centers of SRBFs. Moreover, high-frequency signals are handled by adjusting the bandwidth parameters of SRBFs. To exploit inter-vertex coherence, the light transfer functions are further classified iteratively into disjoint clusters, and tensor approximation is applied within each cluster. Compared with previous methods, the proposed approach enables real-time rendering with comparable quality under high-frequency lighting environments. The data storage is also more compact than previous all-frequency PRT algorithms.

In Chinese landscape painting, rock textures portray the orientation of mountains and contribute to the atmosphere. Many landscape-painting skills are required according to the type of rock. Landscape painting is the major theme of Chinese painting. Over the centuries, masters of Chinese landscape painting developed various texture strokes. Hemp-fiber and axe-cut are two major types of texture strokes. A slightly sinuous and seemingly broken line, the hemp-fiber stroke is used for describing the gentle slopes of rock formations whereas the axe-cut stroke best depicts hard, rocky surfaces. This paper presents a novel method of synthesizing rock textures in Chinese landscape painting, useful not only to artists who want to paint interactively, but also in automated rendering of natural scenes. The method proposed underwrites the complete painting process after users have specified only the contour and parameters.

Roller coasters are an attractive, adventurous and exciting form of entertainment. In this paper, we present a virtual roller coaster system, which includes roller coaster simulation software, a motion platform and a motion control unit. We investigate the track structure of the roller coaster and propose a methodology to generate the track automatically. We also simulate the forces exerted on the running carriage to give the user the impression he/she is riding a real roller coaster. A motion platform with two degrees of freedom is constructed using hydraulic cylinders for its actuators. An electronic circuit is designed to allow the motion platform to communicate with the host computer. We also provide flexibility for changing the track type and the surrounding scenery, in order to give the rider a variety of exciting experiences. This system also considers safety concerns in the design of the motion platform.

Although current photorealistic rendering techniques can produce very impressive images, the rendered objects are often too clean and shiny. Thus, the resulting images look unnatural. This paper proposes a physically‐based model to simulate the appearance of patinas on ancient Chinese bronzes. Buried in the soil for thousands of years, many patinas are found on the surface of ancient bronzes as a result of the aging process and the physical and chemical conditions of the soil environment. The development of patinas is modulated herein by L‐systems according to tendencies based on the environmental factors and object geometry. The tendencies are employed to represent the accumulative effect of all factors on patination. The proposed model can be extended to simulate a variety of metallic patinas including the ancient Chinese bronzes discovered at San‐hsing‐tui, Sichuan, China.

Both numerical and subdivision methods are widely used approaches for ray tracing parametric surfaces. However, the expense of finding the ray–surface intersection points is a major drawback. Thus, simpler and less memory-intensive strategies are needed to improve these methods without further complicating them. This work presents an efficient algorithm for enhancing the performance of both numerical and subdivision methods. The proposed technique can be extended to most applications based on these two methods. The computational time of both approaches is improved by 16–40%. Copyright © 2000 John Wiley & Sons, Ltd.

In this paper we first derive a tighter error bound on form factors as a subdivision criterion for the hierarchical radiosity algorithm. Such an error bound can reduce more unnecessary links and improve the performance of the hierarchical radiosity algorithm to meet a user-specified error tolerance. We then propose a weighted error metric in form factor computation such that more effort is automatically applied to shadow boundaries. Evaluating form factors along shadow boundaries with a higher degree of precision should enhance the quality of human perception. Using the proposed tighter error bound on the weighted error metric, we not only improve the performance but also increase the accuracy of the hierarchical radiosity algorithm. © 1998 John Wiley & Sons, Ltd.

Abstract The texture of feather is one of the most fascinating, complicated, and beautiful texture patterns in nature. In this paper, we propose a new and effective texture generation approach that uses the traits of iteration behavior to synthesize the textures of Galliformes feathers realistically. We also propose an interactive feather modeling approach which provides a close connection between the user's intuition and the resulting branching pattern. In texturing the feather structure, we use an object‐space mapping technique. Experimental results are presented to show the effectiveness of our method.