Digital Human Generation: From 3D Gaussians to Video Generation Large Models
Digital Human Generation: From 3D Gaussians to Video Generation Large Models
Venue: Room A101, School of Computer Science, East Campus, Sun Yat-sen University
Title: Digital Human Generation: From 3D Gaussians to Video Generation Large Models
Speaker: Prof. Yebin Liu, Department of Automation, Tsinghua University
Date: Thursday, May 23, 2024
Time: 10:45 AM – 12:00 PM
Chair: Prof. Weishi Zheng
Abstract: Recently, 3D Gaussian Splatting and the video generation large model Sora have attracted significant attention in the field. As a novel differentiable rendering technique, Gaussian Splatting offers clear advantages in 3D scene representation and rendering due to its higher rendering quality, faster speed, and compatibility with traditional rendering pipelines. Meanwhile, Sora has demonstrated impressive video generation capabilities. Although it is still far from accurately simulating complex real-world scenarios, building and upgrading world simulators will be a key direction for future AI development. Based on the speaker's research in dynamic 3D reconstruction, digital human generation, and dynamic 3D editing, this talk will explore future trends in 3D representation, digital human generation, and world simulation.
Biography of the Speaker:
Prof. Yebin Liu is a tenured professor at the Department of Automation, Tsinghua University, and a recipient of the National Natural Science Foundation of China (NSFC) Excellent Young Scholars Program. His research interests include 3D vision, 3D/4D content generation, digital humans, and holographic communications. He has published over 100 papers in top venues such as TPAMI, SIGGRAPH, CVPR, and ICCV. His technologies have been licensed by eight well-known companies including Huawei, ByteDance, and SenseTime. In the past five years, he has received research funding from leading enterprises such as Huawei, ByteDance, OPPO, Hisense, Apple, Alibaba, Ant Group, Ping An, Kuaishou, BOE, Honor, Lenovo, and SenseTime. He founded and serves as Director of the Tsinghua-Hisense Joint Research Center for Holographic Technologies at the Department of Automation, Tsinghua University. He is an associate editor of IEEE TVCG and CGF, and has served multiple times as Area Chair for CVPR, ICCV, ECCV, and as a member of the Technical Committee of SIGGRAPH Asia. He is also Deputy Director of the CIGS Technical Committee on 3D Vision. He was awarded the National Technology Invention Award (First Prize, Rank #3) in 2012, the China Electronics Society Technology Invention Award (First Prize, Rank #1) in 2019, and received the NSFC Distinguished Young Scholar Grant in 2021.
