Guanbo Bao

Fast Multi-Operator Image Resizing and Evaluation

Current multi-operator image resizing methods succeed in generating impressive results by using image similarity measure to guide the resizing process. An optimal operation path is found in the resizing space. However, their slow resizing speed caused by inefficient computation strategy of the bidirectional patch matching becomes a drawback in practical use. In this paper, we present a novel method to address this problem. By combining seam carving with scaling and cropping, our method can realize content-aware image resizing very fast. We define cost functions combing image energy and dominant color descriptor for all the operators to evaluate the damage to both local image content and global visual effect. Therefore our algorithm can automatically find an optimal sequence of operations to resize the image by using dynamic programming or greedy algorithm. We also extend our algorithm to indirect image resizing which can protect the aspect ratio of the dominant object in an image.

Fast local color transfer via dominant colors mapping

Color transfer is an image editing technique which arises various applications, from daily photo appearance enhancement to movie post-processing. An ideal color transfer algorithm should keep the scene from the source image and apply the color style of the target image. All the dominant colors in the target image should be transferred to the source, while the colors in the source image which are apparently distinct from the target style should not appear in the result. The preservation of the scene details is also important for a good color transfer algorithm.

Low-Resolution Remeshing Using the Localized Restricted Voronoi Diagram

A big problem in triangular remeshing is to generate meshes when the triangle size approaches the feature size in the mesh. The main obstacle for Centroidal Voronoi Tessellation (CVT)-based remeshing is to compute a suitable Voronoi diagram. In this paper, we introduce the localized restricted Voronoi diagram (LRVD) on mesh surfaces. The LRVD is an extension of the restricted Voronoi diagram (RVD), but it addresses the problem that the RVD can contain Voronoi regions that consist of multiple disjoint surface patches. Our definition ensures that each Voronoi cell in the LRVD is a single connected region. We show that the LRVD is a useful extension to improve several existing mesh-processing techniques, most importantly surface remeshing with a low number of vertices. While the LRVD and RVD are identical in most simple configurations, the LRVD is essential when sampling a mesh with a small number of points and for sampling surface areas that are in close proximity to other surface areas, e.g., nearby sheets. To compute the LRVD, we combine local discrete clustering with a global exact computation.

Billboards for Tree Simplification and Real-Time Forest Rendering

Modeling and rendering of large forest scenes are a challenging task, whose geometric complexity surpasses even the capabilities of current graphics hardware. A tree model simplification tool is developed in this paper to automatically generate both hybrid polygon/image-based tree models for close distance viewing and cross billboards for far distance viewing. In the hybrid model, the foliage of the tree is replaced by a set of arbitrarily oriented billboards while the trunk and branches remains polygon models, so that the 3D perception is enhanced and weak parallax of usual image based approaches is avoided. A fast forest rendering system is also developed. We present techniques in the system for rendering forest consisting of many highly detailed plant models. This forest rendering system allows us to render a large forest scene in real-time. Walk-through and flyover a forest are also simulated, illustrating applications for computer games and interactive visualization of landscapes.

Realistic real-time rendering for large-scale forest scenes

Fast rendering of a large-scale forest landscape scene is important in many applications, as video games, Internet graphics applications, landscape or cityscape scene design and visualization, and virtual forestry. A challenge in virtual reality is realistic rendering of large scale scenes consisting of complex plant models. A series of level of detail tree models are usually constructed to compress the overall forest complexity in view-dependent forest navigation. In this paper a new leaf modeling method is presented to have leaf models match leaf textures, so that the visual effect and model complexity can be balanced well. In addition, vertex buffer objects and tree clipping operation allow rendering a large forest containing thousands of trees in real-time. The experiments show that these techniques can be easily used in applications such as video games and interactive navigation of landscapes. Walk-through and flyover a forest are both feasible using our techniques.

Efficient triangulation of Poisson-disk sampled point sets

In this paper, we present a simple yet efficient algorithm for triangulating a 2D input domain containing a Poisson-disk sampled point set. The proposed algorithm combines a regular grid and a discrete clustering approach to speedup the triangulation. Moreover, our triangulation algorithm is flexible and performs well on more general point sets such as adaptive, non-maximal Poisson-disk sets. The experimental results demonstrate that our algorithm is robust for a wide range of input domains and achieves significant performance improvement compared to the current state-of-the-art approaches.

Tree branch level of detail models for forest navigation

We present a level of detail (LOD) method designed for tree branches. It can be combined with methods for processing tree foliage to facilitate navigation through large virtual forests. Starting from a skeletal representation of a tree, we fit polygon meshes of various densities to the skeleton while the mesh density is adjusted according to the required visual fidelity. For distant models, these branch meshes are gradually replaced with semi‐transparent lines until the tree recedes to a few lines. Construction of these complete LOD models is guided by error metrics to ensure smooth transitions between adjacent LOD models. We then present an instancing technique for discrete LOD branch models, consisting of polygon meshes plus semi‐transparent lines. Line models with different transparencies are instanced on the GPU by merging multiple tree samples into a single model. Our technique reduces the number of draw calls in GPU and increases rendering performance. Our experiments demonstrate that large‐scale forest scenes can be rendered with excellent detail and shadows in real time.

Hardware instancing for real-time realistic forest rendering

Real-time rendering of vegetation is important in many applications, such as video games, internet graphics applications, landscape design and visualization. However, the visualization of large-scale forests has always been a challenge not only due to the high geometric complexity but also due to the small batch problem. Moreover, generating real-time shadows for forests will heavily increase the burden. The batch problem is caused by a large number of graphics API draw calls launched in every frame. Normally, rendering a tree model requires at least one graphics API draw call. As a forest usually consists of thousands of trees and the graphics API invocation is a relatively high cost for CPU, the large-scale forest rendering is often CPU bound.

Somatosensory interaction for real-time large scale roaming

Real-time somatosensory roaming in large-scale scene is a challenging problem because of great computation for the scene rendering, the real-time posture recognition and the instant response to the users action. We present a novel somatosensory interactive roaming method in a realistic large-scale forest scene based on Microsoft Kinect. First, we construct a realistic large-scale forest scene based on GPU acceleration. Then for each action of the user, we encode the directions between joints into direction codes, and generate a posture vector which is used to compare with our posture template table to achieve posture recognition. Combining our forest scene with our multi-thread posture recognition, we can somatosensory interactively control roaming direction, velocity, and changes of light and shadow in real-time for the scene with millions of triangles. Our posture recognition algorithm is very fast and can response to the change of the users posture in time. Experiments demonstrate that our method could successfully roam in the large-scale scene with a natural manner, which provides a feasible solution for the challenging problem.

A Survey on Recent Patents in Texture Synthesis

Textures have been a research focus for many years in human perception, computer graphics and computer vi- sion. Recently, research activities in this area emphasize on texture synthesis. Given one or more example textures, a tex- ture synthesis algorithm generates a new one bearing the same visual characteristics. The synthesized texture can be made of arbitrary size specified by the user. It can be used in many fields such as computer animation, virtual reality, data com- pression, non-photorealistic rendering and so on. In this paper, the basic concepts of texture and the development process of texture synthesis techniques are introduced first. Then the recent key patents on texture synthesis schemes are reviewed. Finally, this paper points out future works in this area.