微表情是一种短暂的、微弱的、无意识的面部微表情，持续时间往往在0.5s内，能够揭示人类试图隐藏的真实情绪。微表情识别的研究旨在让机器有足够的智能，能够从人脸视频序列中识别人类的真实情绪。然而由于微表情持续时间短、面部肌肉运动强度低，对其进行准确的表征与识别是一项极具挑战性的任务。为了促进心理学领域和计算机视觉领域针对微表情的进一步研究，由中国图象图形学学会（CSIG）和中国科学院心理研究所举办、CSIG机器视觉专业委员会、CSIG情感计算与理解专业委员会和中科院青促会心理所小组联合承办，中国科学院心理研究所的王甦菁博士和李婧婷博士组织一系列云上微表情的学术活动。
本期活动为一次special session, 邀请到来自英国曼彻斯特城市大学的Moi Hoon Yap教授（SAMM数据库建立者，国际微表情挑战大赛发起人之一）作报告，主题为Translating computer vision and AI into real world applications，欢迎大家关注！
Speaker:
Prof. Moi Hoon Yap
Professor of Image and Vision Computing
Website: https://www.mmu.ac.uk/staff/profile/professor-moi-hoon-yap
Google Scholar: https://scholar.google.co.uk/citations?user=dOSXMNMAAAAJ&hl=en
Biography: Prof. Moi Hoon Yap is the Research Lead in the Department of Computing and Mathematics at Manchester Metropolitan University, UK. Her leadership in research and education has attracted international students and fostered numerous research collaborations. She leads the Human-Centred Computing Group, which consists of 20 staff members and 12 research scholars, and specializes in computer vision and deep learning. As a Royal Society Industry Fellow (2016-2022) in partnership with Image Metrics Ltd, Prof. Yap’s research is informed by industry needs. Her work provides valuable insights and breakthroughs in medical image analysis and facial analysis. She has received research funding from The Royal Society, the EU, EPSRC, Innovate UK, Cancer Research UK, and industry partners. Additionally, she leads technology development for multiple computer vision projects, has created novel datasets to support reproducible research, and has organised international computer vision challenges.
Topic: Translating Computer vision and AI research into real-world applications
Summary: This guest lecture introduces research activities in the Department of Computing and Mathematics at Manchester Metropolitan University, UK, and offers a pathway for potential research collaboration. Prof. Yap will share her research, covering the conceptual foundations and methodologies used in developing medical and computer vision datasets and software over the past decade, illustrated with a timeline to show progress. The lecture will address data-capturing methods, an overview of research efforts in creating private and public datasets, related computer vision tasks (such as facial micro-expression challenges and diabetic foot ulcer challenges), and the future direction of her research.
As a leading institution and investigator in these fields, Prof. Yap aims to share the technical challenges encountered, along with best practices in dataset creation, code/software development, and to inspire other researchers to engage in data sharing within this domain. Future research efforts will focus on forming an international consortium to establish a global repository for medical imaging datasets.
For more details, please refer to:
https://dfu-challenge.github.io/ (DFU challenges)
https://megc2024.github.io/ (Facial micro-expressions Challenges).