On June 30, 2025, Assistant Professor Si Chenyang of our school invited President's Chair Professor Chen Change Loy (Lu Jianqin) from Nanyang Technological University in Singapore to the school to give a report titled From Segment Anything Efficiently to Matting Anyone Precisely.
Abstract：I will present a unified view of recent advances in efficient segmentation and matting, beginning with EdgeSAM - a compact, prompt-aware variant of Segment Anything that achieves real-time segmentation on mobile devices via prompt-in-the-loop distillation. Building on this, EdgeTAM extends promptable segmentation to video with a spatial perceiver for memory compression and a two-stage distillation pipeline, enabling high-quality tracking at 16 FPS on smartphones. Finally, MatAnyone introduces a memory-based framework for video matting, achieving fine boundary details and temporal consistency through region-adaptive memory fusion and segmentation-guided supervision.
Speaker Introduction：Lu Jianqin is a President's Chair Professor at Nanyang Technological University. His research primarily focuses on computer vision and deep learning. To date, he has published over 250 papers in top-tier international academic journals and conferences, including more than 220 papers in CVPR/ICCV/ECCV/NeurIPS/ICLR and SPM/IJCV/TPAMI. Several of his CVPR/ICCV/ECCV papers have been accepted as oral presentations. His papers have been cited over 100,000 times, with an H-index of 133 (according to Google Scholar statistics). He was awarded the Nanyang Scholar Award in 2019 and the Nanyang Research Award in 2024. From 2020 to 2024, he was consistently ranked among the top 100 scholars on the Global 2000 Most Influential Scholars in AI list for five consecutive years. The SRCNN proposed by him and his team is a landmark work in image super-resolution, which has had a significant impact on subsequent research. The relevant paper was selected as one of the Most Popular Articles in TPAMI from March 2016 to the present. This paper also received the Evergreen Paper Award from the Chinese Computer Federation's Computer Vision Committee (CCF-CV). He and his team have also established several landmark computer vision datasets, including the largest and most challenging WIDER FACE database in the field of face detection. He served as an Associate Editor of TPAMI (2019–2024) and CVIU (2023–2025), and is currently an Associate Editor of IJCV. He has served as an Area Chair for top international conferences such as ICCV, CVPR, ECCV, NeurIPS, and ICLR. He is currently the Program Chair for CVPR 2026 and will serve as the General Chair for ACCV 2028.
