Facial Representation Learning & its Applications in Facial Editing

Talk by Ying TAI

Nov 30, 2023 Thursday


Artificial Intelligence Generated Content (AIGC) is leading a profound transformation, reshaping the production of digital content and disruptively accelerating the efficiency of content creation. As a crucial component of AIGC technology, face editing and generation has the great application potential in various practical scenarios, including social entertainment, creative content production, and virtual character creation. Our focus spans several pivotal dimensions: efficiency, controllability, and universality, contextualized within the realm of multi-modal facial generation tailored for real-world environments. This involves the incorporation of diverse modalities such as input audio, text, and images. Specifically, we will provide a brief introduction of representative facial editing and generation tasks, along with the current challenges. Subsequently, we will present our recent works individually, each contributing to advancements in facial editing and generation. Finally, we will provide a comprehensive summary and outline future directions.


Nov 30, 2023 Thursday



RmE1-101, GZ Campus


628 334 1826 (PW: 234567)

Bilibili Live:

ID: 30748067

Speaker Bio:

Asso. Prof. Ying Tai

School of Intelligence Science and Technology, Nanjing University

Ying Tai is an Associate Professor at the School of Intelligence Science and Technology, Nanjing University. Formerly, he served as a principal researcher (T12) and research leader at Tencent Youtu Lab. He earned his bachelor’s and doctoral degrees from Nanjing University of Science and Technology in 2012 and 2017 respectively, advised by Prof. Jian Yang. His primary research focus lies in generative computer vision technologies, encompassing human-centric content editing and generation, high-fidelity image/video restoration, and multi-modal content understanding and generation. With over 80 publications in top conferences and journals, his work has garnered over 9,500 citations on Google Scholar (as of November 2023). He holds over 40 granted patents. His notable achievements include the First Prize of 2021 Progress in Science and Technology of Jiangsu Province (as the fourth contributor) and receiving a Nomination Award of 2018 China Society of Image and Graphics (CSIG) Outstanding Doctoral Thesis. Currently, he serves as the Associate Editor of the Image and Vision Computing journal. He also served as the Area Chairs for ECCV 2022, WACV 2024 & 2023, and FG 2023. Moreover, he has been recognized as one of the annual Top 2% Scientists by Stanford University from 2020 to 2022.