关于举办王浩博士学术报告会的通知

发布时间:2023-09-25 浏览次数:424

报告题目:StyleGAN-based Visual Content Generation

报 告 人:王浩博士

主 持 人:吴庆耀教授                           

报告时间:2023926日 (星期二)10:00—12:00

线下参加:华南理工大学大学城校区B7-339

线上参加:腾讯会议:123-736-266                                                  

欢迎广大师生踊跃参加!


摘要:

Generating 2D and 3D visual content is a significant task in generative AI. One of the key challenges in this domain is generating high-quality results while maintaining efficiency. In this talk, I will introduce our recent work in the area of visual content generation, covering topics such as 2D face manipulation, 3D point cloud generation, 3D cartoon face generation and text-to-3D object generation. The connections between 2D and 3D generation technology based on the StyleGAN model architecture will also be presented


个人简介:

王浩博士,香港科技大学(广州)人工智能学域助理教授,2023年博士毕业于新加坡南洋理工大学计算机科学与工程学院,师从Miao ChunyanLin GuoshengSteven Hoi教授。他的研究兴趣在于为多模态领域开发AI驱动的感知和生成算法,例如视觉与文本数据之间的转化。此前曾是TikTok、地平线和中科院自动化所的实习生。以第一作者身份在CVPR, ECCV, ACM-MM, IEEE TPAMI, IEEE TIP, Pattern Recognition, IEEE TMM等顶级计算机视觉、多媒体会议上发表十余篇论文。