GAMES Webinar 2024 – 340期(智能CAD模型生成) | 吴其兵(山东大学),余锋根(Simon Fraser University)

 

【GAMES Webinar 2024-340期】(几何专题-智能CAD模型生成)

报告嘉宾:吴其兵(山东大学)

报告时间:2024年9月19号星期四晚上8:00-8:25(北京时间)

报告题目:Tune-It Optimizing Wire Reconfiguration for Sculpture Manufacturing

报告摘要:

该论文的研究主题是复杂几何线制品的成型制造问题。线制品广泛应用于工业生产和日常生活,如线艺术雕塑、弹簧、金属丝栅栏等。弯线机在线制品的数控加工方面具备高效率和高精度的优点,但受制于免碰撞的约束,只能生产结构简单的线制品,尚不支持复杂几何线制品的成型制造。本文提出了一种人机协同的两阶段成型制造计算框架,第一阶段中,使用弯线机制造一个与目标形状局部相同且免于碰撞的线制品形状;第二阶段中,由人工调整少数几个节点的弯曲程度,即可得到目标线制品。最终,该方法较好的解决了复杂几何线制品的制造问题。

讲者简介:

吴其兵,山东大学计算机学院交叉研究中心三年级硕士生,指导老师为屠长河教授和赵海森教授,研究兴趣为计算机图形学和智能制造。目前以共同第一作者身份在SIGGRAPH Asia 2024发表论文一篇。

讲者主页:https://qibingwu.github.io/


报告嘉宾:余锋根(Simon Fraser University)

报告时间:2024年9月19号星期四晚上8:25-9:10(北京时间)

报告题目:Learning Structured Representations of 3D CAD Models

报告摘要:

Computer-Aided-Design (CAD) models have become widespread in engineering and manufacturing, driving decision-making and product evolution related to 3D models. Understanding the structure of 3D CAD models is crucial for various applications, as it can significantly benefit 3D shape analysis, modeling, and manipulation. In this talk I will talk about my past works about learning structured representations of 3D CAD models, including part-set representation, CSG representation and hierarchical representation. We have proposed a series of deep-learning-based techniques that enable the reconstruction of structured 3D CAD models, using diverse input data such as voxels, point clouds, and images. In the end, I will discuss how recent large foundation models can benefit learning the structured representations of 3D CAD models.

讲者简介:

Fenggen Yu recently obtained his Ph.D. from Gruvi Lab, school of Computing Science at Simon Fraser University and will join Amazon as research scientist. He received his bachelor degree and master degree from Nanjing University. His research interests are on 3D Geometric Modeling and Analysis, Geometric Deep Learning, 3D Content Generation, and 3D Reconstruction from multi-sensors. His recent works were published on CVPR, ICCV, ECCV, NeurIPS, SIGGRAPH Asia and TOG.

讲者主页:https://fenggenyu.github.io/index.html


主持人简介:

朱晨阳,博士,国防科技大学计算机学院计算科学系副教授,研究方向为计算机图形学,三维视觉及几何形状分析。入选第六届中国科协青年托举计划,湖南省湖湘青年英才计划。任中国工业与应用数学学会几何设计与计算专委会副秘书长。共发表CCF A类论文及SCI 1区论文30余篇,其中计算机图形学顶级会议SIGGRAPH/TOG文章6篇,计算机视觉顶级会议CVPR Oral 1篇,AAAI Oral 1篇,ESI引用进入1%的论文2篇。研究成果“流形约束保结构建模”获得湖南省自然科学一等奖1项(排名4)。主讲课程获军队级金课1门(排名2),湖南省金课1门(排名5)。担任计算机图形学国际刊物The Visual Computer,国内核心期刊《图学学报》等期刊编委,EG,PG,AAAI, IJCAI, CGI, CVM等会议程序委员。


GAMES主页的“使用教程”中有 “如何观看GAMES Webinar直播?”及“如何加入GAMES微信群?”的信息;
GAMES主页的“资源分享”有往届的直播讲座的视频及PPT等。
观看直播的链接:https://live.bilibili.com/h5/24617282

You may also like...