项目名称: 基于内容的三维动画数据检索
项目编号: No.61472223
项目类型: 面上项目
立项/批准年度: 2015
项目学科: 计算机科学学科
项目作者: 彭京亮
作者单位: 山东大学
项目金额: 82万元
中文摘要: 随着三维动画的广泛应用以及数据量的飞速增长,快速、方便、准确、智能化的三维动画数据检索就成为急需解决的问题,同时也成为一个巨大的挑战。 然而,我们尚未发现有专门的三维动画数据检索方面的研究成果发表。 在本项目中,我们将率先对于基于内容的三维动画数据检索进行研究,着重解决面向三维动画数据的特征描述、数据匹配和交互式查询定义这三个方面的诸多技术难题,从而形成多种有效的问题解决方案。 本申请书所提出的一系列问题解决方案中,有的基于低层数据特征的分析和描述,有的基于中层或者高层数据特征的分析和描述,有的二者兼顾。不同的解决方案各有其优缺点,可以适用于不同的应用需求。 本项目的研究目标是在著名国际会议或者期刊中发表相关学术论文8-10 篇,其中包括顶级学术会议或者期刊论文3-5 篇。
中文关键词: 基于内容的检索;三维动画;运动捕捉
英文摘要: Three-dimensional (3D) animations are becoming more and more widely used and, at the same time, the data volume of 3D animations is explosively increasing. As such, efficient, convenient, accurate and intelligent data retrieval has become an urgent problem and a big challenge at the same time. However, we have not seen publications specifically on 3D animation retrieval yet. In this project, we will make the first steps in researching content-based 3D animation data retrieval. We will focus on key technical problems in three aspects, feature description, data matching and interative query specification, leading to various effective solutions. Of the proposed solutions, some are based on low-level data feature analysis and description, some are based on middle- or high-level data feature analysis and description, while some others are based on both. Different solutions have their specific advantages and disadvantages and potentially meet different application demands. The goal of this project is to publish 8-10 papers in renowned international conferences' proceedings or journals, which include 3-5 papers published in top-notch conferences' proceedings or journals.
英文关键词: Content-Based Retrieval;3D Animation;Motion Capture