基于双目视觉的动画图像三维建模方法  

3D Modeling Method for Animated Images Based on Binocular Vision

作　　者：买欣蕾 任山山 MAI Xinlei;REN Shanshan(Department of Basic Education,Huanghe Jiaotong University,Jiaozuo Henan 454950,China)

机构地区：[1]黄河交通学院基础教学部,河南焦作454950

出　　处：《信息与电脑》2023年第9期34-36,共3页Information & Computer

摘　　要：针对现有建模方法应用效果不佳的问题,提出一种基于双目视觉的动画图像三维建模方法。首先,进行动画图像特征提取及数据采集,设定基础建模点位多层级;其次,建立纹理映射建模关系,构建双目视觉三维图像结构,通过Java3D识别完成三维建模;最后,进行对比分析。测试结果表明,应用该方法得出的像点偏差被较好地控制0.2以下,具有实际应用价值。A three-dimensional modeling method for animated images based on binocular vision is proposed to address the issue of poor application performance of existing modeling methods.Firstly,feature extraction and data collection of animated images are carried out,and basic modeling points are set at multiple levels.Secondly,establish texture mapping modeling relationship,construct binocular vision 3D image structure,and complete 3D modeling through Java 3D recognition.Finally,conduct a comparative analysis.The test results show that the pixel deviation obtained by applying this method is well controlled below 0.2,which has practical application value.

关 键 词：双目视觉 动画图像 三维建模 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]