【预告】CSIG图像图形技术国际在线研讨会第七期香港专场（图像视频分析专题）将于8月4日举办

2022 年 7 月 30 日 中国图象图形学学会CSIG

为了响应国家号召，加强粤港澳大湾区科技发展与交流工作的协同共进，值此庆祝香港回归祖国25周年之际，图像图形技术国际在线研讨班将于8月4日举办香港专场，邀请香港科技大学徐旦博士、香港中文大学窦琪博士、香港城市大学李皓亮博士围绕“图像视频分析”主题开展学术交流活动，分享最新前沿技术、搭建香港-内地青年学者的交流平台，共话粤港澳科技协同发展的美好蓝图。期待学术与工业界同行的积极参与！参会老师和同学可以直接进入腾讯会议号参加，并同步在CSIG视频号和蔻享学术直播。

主办单位： 中国图象图形学学会（CSIG）

承办单位： CSIG国际合作与交流工作委员会、女科技工作者工作委员会、青年工作委员会

会议时间： 2022年8月4日（周四）14:00-17:00

腾讯会议号： 243-784-845

直播链接：

会议日程

讲者简介

Dr. Dan Xu

Dr. Dan Xu is an Assistant Professor in the Department of Computer Science and Engineering (CSE), Hong Kong University of Science and Technology (HKUST). Before joining HKUST, he was a Postdoctoral Researcher in the Visual Geometry Group (VGG) at the University of Oxford, working with Prof. Andrea Vedaldi and Prof. Andrew Zisserman. He received his Ph.D. in Computer Science from the University of Trento in 2018, under the supervision of Prof. Nicu Sebe. He was also a visiting Ph.D. student in the MMLab at the Chinese University of Hong Kong (CUHK) under the supervision of Prof. Xiaogang Wang. His research mainly focuses on computer vision, multimedia, and deep learning. Specifically, he is interested in multi-modal and structured representation learning, statistical modeling within deep learning, as well as their applications in 2D/3D scene understanding. He served as Senior Programme Committee (SPC) / Area Chair (AC) at multiple international conferences including AAAI, ACM Multimedia, WACV, and ACCV. He received the Best Scientific Paper Award at ICPR 2016 and a Best Paper Nominee at ACM Multimedia 2018.

Talk title：Multi-Modal Multi-Task 2D/3D Scene Understanding with Least Efforts of Annotations

Abstract: The talk will cover several important computer vision tasks within the context of visual 2D/3D scene understanding, including scene depth estimation, joint learning of scene depth and scene parsing, and visual deep SLAM. Specifically, the talk will present how to design effective deep learning models to deal with these challenging problems while using the least efforts of human annotations, including the introduction of a transformer network structure for deep multi-task learning, a weakly supervised model for scene parsing, and self-supervised joint learning of scene depth, camera pose, and moving object discovery. At the end of the talk, several current research directions in our Lab will also be introduced.

Dr. Qi Dou

Dr. Qi DOU is an Assistant Professor with the Department of Computer Science and Engineering at The Chinese University of Hong Kong. She is also an Associate Member of the T Stone Robotics Institute and Multi-Scale Medical Robotics Center at CUHK. Her research focuses on synergistic innovations across medical image analysis, machine learning, surgical data science and medical robotics, with an impact to support demanding clinical workflows for improving patient care. Dr. Dou has won the IEEE EMBS TBME Best Paper Award 2nd Place 2021, IEEE ICRA Best Paper Award in Medical Robotics 2021, MICCAI Young Scientist Publication Impact Award Finalist 2021, MICCAI-Medical Image Analysis Best Paper Award 2017. Dr. Dou serves as the associate editor for Journal of Machine Learning for Biomedical Imaging, Computer Assisted Surgery, and serves as program co-chair of MICCAI 2022, MIDL 2022&2021, IPCAI 2023.

Talk title: Video-based Robotic Surgery Intelligence

Abstract: With rapid advancements in medicine and engineering technologies, the operating room has evolved to a highly complex environment, where surgeons take advantages of computers, endoscopes and robots to perform procedures with more precision while less incision. Intelligence, together with its authorized cognitive assistance, smart data analytics, and automation, is envisaged to be a core fuel to transform next-generation robotic surgery in numerous known or unknown exciting ways. In this talk, I will present ideas, methodologies and applications of image-based robotic surgery intelligence from three perspectives, i.e., Al-enabled surgical situation awareness to improve surgical procedures, Al-powered large-scale data analysis to enhance surgical education, Al-driven multi-sensory perception to achieve surgical subtask automation. To tackle these challenging topics, a wide variety of cutting-edge vision and learning techniques will be covered, including transformers, weakly-supervised learning, meta-learning model generalization, unsupervised video retrieval, stereo depth estimation, 3D scene reconstruction, reinforcement learning and augmented reality.

Dr. Haoliang Li

Dr. Haoliang Li received the B.S. degree in communication engineering from University of Electronic Science and Technology of China (UESTC) in 2013, and his Ph.D. degree from Nanyang Technological University (NTU), Singapore in 2018. He is currently an assistant professor in Department of Electrical Engineering, City University of Hong Kong. His research mainly focuses on AI security, multimedia forensics and transfer learning. His research works appear in international journals/conferences such as TPAMI, IJCV, TIFS, ICML, NeurIPS, and CVPR. He received the ACM SIGSOFT Distinguished Paper award, Wallenberg-NTU presidential postdoc fellowship, doctoral innovation award, and VCIP best paper award, and ACM SIGSOFT distinguished paper award.

Talk title: Towards Better Generalized Visual Recognition Algorithms for Real-World Challenges

Abstract: Visual recognition tasks have been greatly benefited from the fast developments of deep learning approaches. However, the success of deep learning lies in the strong assumption that the training and test data are drawn from similar feature spaces and distributions, which may not always be valid in realistic scenarios. As such, the existing deep learning-based techniques lack generalization capability which further prohibits themselves from being applied to different practical applications. In this talk, I will introduce several deep learning-based domain generalization approaches to tackle the problem of generalization capability. The introduced methods are based on fundamental studies in the field of machine learning (distribution alignment, variational Bayesian) and software engineering (coverage testing). I will also discuss some possible future directions for the generalization issues of deep learning algorithms.

主席

丛润民

丛润民，北京交通大学信息科学研究所、数字媒体信息处理研究中心副教授，硕士生导师，入选中国科协“青年人才托举工程”、“北京市科技新星”计划。主要研究方向包括计算机视觉、多媒体信息处理、注意力感知与显著性计算、遥感影像解译与分析、开放环境下视觉内容增强等。在TIP、TCyb、TII、TMM、TCSVT、TGRS、NeurIPS、CVPR、ECCV、AAAI、IJCAI、ACM MM等国内外学术期刊及会议上发表论文60余篇，其中CCF A/IEEE Trans 论文42篇，ESI热点论文2篇、高被引论文9篇；出版英文专著章节2部。担任Neurocomputing、IEEE Journal of Oceanic Engineering等SCI期刊编委。荣获IEEE ICME最佳学生论文奖亚军、天津市科学技术进步一等奖、中国图象图形学学会优秀博士学位论文奖（全国10篇）、IEEE CVPR杰出审稿人、第十五届北京青年优秀科技论文奖、北京图象图形学学会优秀博士学位论文奖（京津冀6篇）等。

许倩倩

许倩倩，中科院计算所副研究员，博士生导师，国家优秀青年基金获得者。IEEE/CSIG/CCF高级会员，CSIG青工委副秘书长、CSIG多媒体专委会副秘书长，CAAI深度学习专委会副秘书长。研究领域为数据挖掘和机器学习，主要关注群智计算和知识图谱，已在TPAMI、IJCV、TIP、TKDE、ICML、NeurIPS、CVPR、AAAI、ACM Multimedia等国际期刊和会议上发表CCF-A类论文60余篇。先后获得：吴文俊人工智能自然科学奖一等奖，中国人工智能学会最佳青年科技成果奖，中国图象图形学学会石青云女科学家奖，吴文俊人工智能优秀青年奖，ACM中国SIGMM新星奖, 中国人工智能学会优秀博士学位论文，中科院百篇优秀博士学位论文，CCF-腾讯犀牛鸟科研基金、首届CAAI-华为MindSpore学术奖励基金等奖励。担任国际期刊TMM、T-CSVT、ACM TOMM和Multimedia Systems编委，CCF-A类国际会议ACM MM领域主席，AAAI和IJCAI SPC。

来源：CSIG国际合作与交流工委会