Modern image processing techniques increasingly use prior models of the expected distribution of objects. Principal component eigen-models are often selected for shape prior modeling, but are limited in capturing only...
详细信息
Modern image processing techniques increasingly use prior models of the expected distribution of objects. Principal component eigen-models are often selected for shape prior modeling, but are limited in capturing only the second order moment statistics. On the other hand, kernel densities can in concept reproduce arbitrary statistics, but are problematic for high dimensional data such as shapes. An evident approach is to combine these methods, using PCA to reduce the problem dimensionality, followed by kernel density modeling of the PCA coefficients. In this paper we show that useful algorithmic and editing operations can be formulated in term of this simple approach. The operations are illustrated in the context of point distribution shape models. Particular points can be rapidly evaluated as being plausible or outliers, and a plausible shape can be completed given limited operator input in a manually guided procedure. This "PCA+KD" approach is conceptually simple, scalable (becoming increasingly accurate with additional training data), provides improved modeling power, and supports useful algorithmic queries.
This paper presents an interactive technique for the dense texture-based visualization of unsteady 3D flow, taking into account issues of computational efficiency and visual perception. High efficiency is achieved by ...
详细信息
Shadow maps are a widely used shadowing technique in real time graphics. One major drawback of their use is that they cannot be filtered in the same way as color textures, typically leading to severe aliasing. This pa...
详细信息
We present a vector graphics representation suitable for real-time rendering on CPUs. Our representation can be used in place of a texture map, and renders precise antialiased edges at any magnification. A combination...
详细信息
In this paper, we propose a multiresolution framework for improving the quality of several image and audio processing algorithms. The results of algorithms operating at different time-frequency (or space-frequency) re...
详细信息
In this paper, we propose a multiresolution framework for improving the quality of several image and audio processing algorithms. The results of algorithms operating at different time-frequency (or space-frequency) resolutions are adaptively combined in order to achieve a variable resolution of a filter bank. Applications of the proposed model to image noise reduction algorithms are demonstrated with examples of non-local means and adaptive PCA algorithms
作者:
Lele ZhouSaif ZahirImage Processing
Graphics and Multimedia Lab Computer Science Department University of Northern British Columbia Canada
Binary image compression is desirable for a wide range of applications, such as digital libraries, map archives, fingerprint databases, facsimile, etc. In this paper, we present a new highly efficient algorithm for lo...
详细信息
Binary image compression is desirable for a wide range of applications, such as digital libraries, map archives, fingerprint databases, facsimile, etc. In this paper, we present a new highly efficient algorithm for lossless binary image compression. The proposed algorithm introduces a new method, direct redundancy elimination, to efficiently exploit the two-dimensional redundancy of an image, as well as a novel dynamic context model to improve the efficiency of arithmetic coding. Simulation results show that the proposed algorithm has comparable compression ratio to JBIG standard. In many cases, the proposed algorithm outperforms the JBIG standard
Demosaicking is one of the important tasks in the image-processing pipeline in digital cameras using a single electronic sensor overlaid with a Color Filter Array. We quantitatively shows that, demosaicking algorithms...
详细信息
An optimal recovery based neural-network Super Resolution algorithm is developed. The proposed method is computationally less expensive and outputs images with high subjective quality, compared with previous neural-ne...
详细信息
We present a simplification algorithm for manifold polygonal meshes of plane-dominant models. Models of this type are likely to appear in man-made environments. While traditional simplification algorithms focus on gen...
详细信息
We present a simplification algorithm for manifold polygonal meshes of plane-dominant models. Models of this type are likely to appear in man-made environments. While traditional simplification algorithms focus on generality and smooth meshes, the approach presented here considers a specific class of man-made models. By detecting and classifying edge loops on the mesh and providing a guided series of binary mesh partitions, our approach generates a series of simplified models, each of which better respects the semantics of these kinds of models than conventional approaches do. A guiding principle is to eliminate simplifications that do not make sense in constructed environments. This, coupled with the concept of "punctuated simplification", leads to an approach that is both efficient and delivers high visual quality. Comparative results are given.
Recent developments in the area of multimedia middleware have proven to provide suitable abstractions for supporting various ubiquitous computing scenarios. As a typical example, home entertainment applications aim at...
详细信息
ISBN:
(纸本)0889865108
Recent developments in the area of multimedia middleware have proven to provide suitable abstractions for supporting various ubiquitous computing scenarios. As a typical example, home entertainment applications aim at providing a "follow-me" service, where audio/video output is performed using devices embedded in the current surrounding of the user. However, manually setting up such distributed multimedia applications is a complex task. Therefore, middleware needs to provide further guidance. In this paper, we provide an in-depth analysis of the challenges for such a service. Based on these observations, an algorithm is derived that allows for the automatic creation of distributed multimedia data flow graphs from a given high-level description. The applicability and performance of our approach is evaluated by demonstrating relevant application scenarios.
暂无评论