Ranking with uncertain labels and its applications

The techniques for image analysis and classification generally consider the image sample labels fixed and without uncertainties. The rank regression problem studied in this paper is based on the training samples with uncertain labels, which often is the case for the manual estimated image labels. A core ranking model is designed first as the bilinear fusing of multiple candidate kernels. Then, the parameters for feature selection and kernel selection are learned simultaneously by maximum a posteriori for given samples and uncertain labels. The provable convergency Expectation Maximization (EM) method is used for inferring these parameters in an iterative manner. The effectiveness of the proposed algorithm is finally validated by the extensive experiments on age ranking task and human tracking task. The popular FG-NET and the large scale Yamaha aging database are used for the age estimation experiments, and our algorithm outperforms those state-of-the-art algorithms ever reported by other interrelated literatures significantly. The experiment result of human tracking task also validates its advantage over conventional linear regression algorithm. A short version of this paper appeared in ICME07.     

Adding Support for Persistence to CLOS via Its Metaobject Protocol

Language-level support for object persistence frees programmers from having to confront a broad class of database issues from within their applications. By virtue of its metaobject protocol, CLOS is a language whose semantics can be tailored by individual programmers. We used the metaobject protocol to extend CLOS with support for object persistence. Our goal was to obtain a version of CLOS with persistence to which we could easily port a commercial geometric CAD modeling system. We describe the design and implementation of our persistence extension and highlight the strengths and weaknesses exhibited by the CLOS metaobject protocol during our experiment. For many aspects of the implementation we found that the metaobject protocol was ideal. In other cases we had to choose among paying a large performance penalty, extending the protocol, and bypassing the protocol by modifying the language implementation directly.     

Mutual information based registration of multimodal stereo videos for person tracking

Research presented in this paper deals with the systematic examination, development, and evaluation of a novel multimodal registration approach that can perform accurately and robustly for relatively close range surveillance applications. An analysis of multimodal image registration gives insight into the limitations of assumptions made in current approaches and motivates the methodology of the developed algorithm. Using calibrated stereo imagery, we employ maximization of mutual information in sliding correspondence windows that inform a disparity voting algorithm to demonstrate successful registration of objects in color and thermal imagery. Extensive evaluation of scenes with multiple objects at different depths and levels of occlusion shows high rates of successful registration. Ground truth experiments demonstrate the utility of the disparity voting techniques for multimodal registration by yielding qualitative and quantitative results that outperform approaches that do not consider occlusions. A basic framework for multimodal stereo tracking is investigated and promising experimental studies show the viability of using registration disparity estimates as a tracking feature.     

A Generic Pigment Model for Digital Painting

We propose a generic pigment model suitable for digital painting in a wide range of genres including traditional Chinese painting and water-based painting. The model embodies a simulation of the pigment-water solution and its interaction with the brush and the paper at the level of pigment particles; such a level of detail is needed for achieving highly intricate effects by the artist. The simulation covers pigment diffusion and sorption processes at the paper surface, and aspects of pigment particle deposition on the paper. We follow rules and formulations from quantitative studies of adsorption and diffusion processes in surface chemistry and the textile industry. The result is a pigment model that spans a continuum from the very wet to the very dry brush stroke effects. We also propose a new pigment mixing method based on machine learning techniques to emulate pigment mixing in real life as well as to support the creation of new artificial pigments. To experiment with the proposed model, we embedded the model in a sophisticated digital brush system. The combined system exhibits interactive speed on a modest PC platform. http://www.cs.hku.hk/~songhua/pigment provides supplementary materials for this paper.     

一种识别薄板对接接头的视觉方法

对薄板紧密对接接头的识别问题进行了深入的分析，提出了由照明、自适应局部特征提取、全局特征识别和验证融合在一起的视觉识别方法。它克服了电弧光的干扰，识别结果可靠，可以有效地解决薄板紧密对接接头识别这一难题。     

The operational semantics of user-defined relationships in object oriented database systems

In semantic data models, abstract relationship (e.g. generalization, aggregation, etc.) semantics are defined, specifying how insertion, deletion and modification operations made at a higher level of abstraction can affect the objects abstracted over and vice versa. These semantics, also known as structural constraints, are expressed through so-called update rules. This perspective has been somewhat lost in most object-oriented systems, where user-defined relationships are supported as simple pointers and their semantics are embedded, distributed and replicated within the operations accessing these pointers. This paper inherits and extends the treatment of relationships found in semantic data models to behavioural object-oriented models by presenting an approach to uniformly capture the update rules for user-defined relationships. The stress is not on supporting relationships as first-class objects, but on describing their update rules (or operational semantics) through a set of constructors namely, reaction, anticipation, delegation and exception. The approach has been borne out by an implementation in an active object-oriented database system.     

多坐标系下目标跟踪的联合滤波

针对雷达目标观测和处理在不同的坐标系下完成，本文提出了联合滤波算法来跟踪机动目标。该算法以卡尔曼滤波器为基础，直角坐标系下和极坐标系下的算法相联合，不仅克服了两种坐标系下滤波算法的不足，而且对机动目标有很好的跟踪效果。仿真实验结果表明了该算法的有效性。     

基于奇异值分解的特征跟踪方法

在传统的基于模板匹配的跟踪方法中，均是给定一个模板，然后从图像中各个位置取出一个个与模板大小一致的区域进行相似性度量，找出与模板距离最小的一个区域作为当前模板，以便进行下一步的匹配跟踪工作。在景象匹配和相关跟踪过程中，由于所面临的大多数是变化的场景，实时获取的图像与预存模板之间存在比较大的差异，传统相关匹配方法的应用就会受到限制；而且在跟踪过程中，随时更新模板会造成跟踪性能对扰动过分敏感，从而产生漂移。首先拍摄目标不同角度的图像（尽可能包含目标可能出现的所有情况），构成目标图像训练集合，抽取出特征矩阵，对它进行奇异值分解，构成一个关于目标的多维空间。然后再用匹配方法在全局范围搜索，找出目标的大致位置，并利用收敛方法在确定的大致位置内进行搜索，确定目标的仿射变换系数，从而得到一个目标位置的确切描述。     

强激光热斑形成及其稳定性研究

在论述热点跟踪及其传热机理时将靶材的比热溶、热传导系、在激光的热藕合系数均考虑成温度和时间的函数，又考虑靶材在运动时将激光辐照物体时热点的建立及其稳定情况作一动态数值模拟，为热点跟踪作准备。     

一种基于Google Maps的土地利用信息发布平台

目前常见的Web地图服务仅支持浏览和查询,不支持矢量数据的发布.文章提出一种基于Google Maps的WebGIS构建方法,支持多种浏览器的矢量数据互操作及空间分析.针对郑州市土地利用信息发布需求,采用B/S架构,借助Google Maps API和AJAX技术,实现了土地利用信息系统化,标准化发布平台.该平台响应速度快,面向多种用户,支持数据共享,能更好的为土地利用变化研究及土地整理决策提供服务.