Seminar by Professor Yunchao Wei from Beijing Jiaotong University, 14 Aug 2024, LT7 North Spine, NS1-02-03

Time: 15:00-16:30, August 14th Wednesday, 2024

Venue: LT7 North Spine, NS1-02-03 (LEVEL 2, BESIDE CEE)

Title: Visual Perception Meets Imperfect Data

Abstract:

"Imperfect" annotated visual media data is widely prevalent, manifested in the following aspects: 1) Difficulty in annotating data for some visual perception tasks, resulting in insufficient annotated samples; 2) Discrepancies in annotated categories between different time periods, leading to conflicts between past and current annotations; 3) Real-world data categories are open-ended, but existing datasets often limit annotations to fixed category sets. To address the issues of insufficient annotated samples, temporal conflicts in annotated categories, and a limited number of categories, Professor Wei has conducted research on some mainstream visual perception tasks (e.g., visual segmentation, referring comprehension, retrieval, etc.). His research outputs were published in recent top international conferences/journals such as NeurIPS, CVPR, ICCV, ECCV, TPAMI.

Bio:

Yunchao Wei is currently a full professor at Beijing Jiaotong University (BJTU). Before join BJTU, he conducted research at the National University of Singapore, the University of Illinois at Urbana-Champaign, and the University of Technology Sydney. He was selected as MIT TR35 China by MIT Technology Review in 2021, Global Outstanding Chinese Young Scholar by Baidu in 2021, and was named as one of the five top early-career researchers in Engineering and Computer Sciences in Australia by The Australian in 2020. He has received numerous awards, including the First Prize of the Ministry of Education Natural Science Award (2022), First Prize of the Science and Technology Award from the China Society of Image and Graphics (2019), Young Researcher Award from the Australian Research Council (2019), IBM C3SR Best Research Award (2019). He received many competition prizes from CVPR/ICCV/ECCV such as the Winner prizes of ILSVRC 2014, LIP 2018/2019, Youtube VOS 2021, Runner-up Prizes of ILSVRC 2017, DAVIS 2020, etc. He has published over 100 papers in top journals/conferences such as TPAMI and CVPR, with over 22,000 citations on Google Scholar. His current research focuses on visual perception with imperfect data, multimodal data analysis, generative artificial intelligence, and more.