项目名称: 少数民族特色视觉艺术的云南重彩画风格化绘制及科学理解研究
项目编号: No.61271361
项目类型: 面上项目
立项/批准年度: 2013
项目学科: 无线电电子学、电信技术
项目作者: 普园媛
作者单位: 云南大学
项目金额: 70万元
中文摘要: 风格化绘制让计算机以更艺术、更反映人主观意识的方式来表达客观世界,在多媒体信息处理领域具有较大的研究和应用价值。云南重彩画是少数民族特色视觉艺术的杰出代表,融合中西方绘画艺术特色,在国内外享有盛誉。本项目针对其鲜明的中国线条画和西方油画特点,分析归纳其在构图、线条、纹理及色彩使用上的特点和规律,探索、研究相应的风格化绘制核心算法。主要研究内容包括重彩画人物形体绘制特点学习和归纳、基于3D人体模型的2D人物变形、重彩画白描图绘制,特有纹理绘制和云南少数民族织绣纹样设计等多项关键技术。同时开展视觉艺术科学理解的研究,基于统计和信息处理的方法,分析、提取、归纳反映画派风格的特征,对其进行定量的数字化描述,并在此基础上探索风格化绘制效果客观评价的途径。项目研究内容可推广到其他视觉艺术流派的风格化绘制和科学理解,研究成果可用于影视制作、动漫设计、画作真伪鉴定等行业,促进了多媒体信息处理技术的发展。
中文关键词: 视觉艺术风格化绘制;视觉艺术科学理解;云南民族视觉艺术;照片美感品质评价;深度学习
英文摘要: Stylistic rendering can be used to depict the objective world by computer in the way of more artistry and more subjective consciousness and has values of researches and applications for multimedia information processing. The Yunnan Heavy Color Painting, which enjoys a high reputation at home and abroad, is an outstanding delegate of minority visual arts and full of minority characteristics. In this project, the key algorithms of Yunnan heavy color painting stylistic rendering will be proposed. Yunnan heavy color painting has characteristics of Chinese line paintings and Western oil paintings. According to this, the characteristics and regularities of the composition, lines, textures and colors of it are analyzed and concluded and computer rendering algorithms are researched respectively. Main research contents include learning and concluding characteristics of Yunnan heavy color painting features, a two-dimensional image deformation based on three-dimensional human modeling, rendering of Yunnan heavy color painting line drawing, rendering of specific textures, digital design of Yunnan minority weaving embroidering dermatoglyphic patterns and so on. Other topics about visual arts scientific understanding are also investigated. Based on the statistical and information processing methods, we will analyze, extract a
英文关键词: Visual art stylistic rendering;Scientific understanding of the visual art;Yunnan minority visual art;images aesthetic quality evaluation;Deep learning