Open Data Base: Open Platform for Multimodal Data Collection from Wearable Devices
August 22 (Friday) 3:30 PM-5:30 PM
Location: Shenyang Hall, 3rd Floor

Introduction:
       Wang Chongyang is an Assistant Research Fellow and Deputy Director of the Smart Rehabilitation Laboratory, Sichuan Key Laboratory of Rehabilitation Medicine, West China Hospital, Sichuan University. He previously completed a postdoctoral fellowship at the Ministry of Education Key Laboratory of Ubiquitous Computing at Tsinghua University and was selected for the Shuimu Scholar Program, with Associate Professor Yu Chun as his co-supervisor. He received his PhD from the University College London Human-Computer Interaction Centre in 2022, receiving a full UCL PhD scholarship (UCLORS & GRS), with Professor Nadia Berthouze as his supervisor. His research focuses on smart rehabilitation, focusing on natural perception and interaction with human movement, and interactive embodied intelligence based on mobile multimodal robots. He has published numerous papers at CCF-A international conferences on human-computer interaction, ubiquitous computing, and artificial intelligence (CHI, Ubicomp, AAAI, and IJCAI), as well as international academic journals such as IEEE T-Ro, NeuroImage, and ACM Health. He was selected for the National Postdoctoral Overseas Talent Recruitment Program and received the CCF TPCI 2023 Best Paper Reviewer Award. He is also the lead researcher for projects funded by the National Natural Science Foundation of China (Class C) Young Scientist Fund and other postdoctoral fellowships.

Report Title: Practice and reflection on innovative wearable devices in the intersection of clinical medicine and engineering

Report Introduction:  

        In recent years, previous-generation wearable devices, exemplified by smart bracelets and watches, have demonstrated significant research value in clinically-focused randomized controlled trials. Researchers in the fields of human-computer interaction and ubiquitous computing have optimized the intelligence of these devices through data modeling and analysis, while medical professionals have leveraged these technologies to enable dynamic monitoring, predictive classification, and precise intervention for related diseases. However, as the low-hanging fruit in this field gradually depletes, researchers urgently need to explore more innovative research paths. This presentation, drawing on the speaker's recent experience, will explore two key areas: 1. Innovative applications of common devices—exploring the potential clinical value of traditional wearable devices; and 2. Exploring new devices—promoting the deep integration of cutting-edge technologies with medical needs. Through case studies and empirical insights, the speaker will attempt to propose potential pathways for the integration of wearable devices and medicine, and delve into the technical bottlenecks, clinical challenges, and potential solutions faced in this research process.

Introduction:
       As the head of AI identity at Ant Group, his research interests include human-computer interaction, identity security, computer vision, multimodal understanding, and biometrics. He has published over 20 papers in top international conferences and journals, including NeurIPS, CVPR, ICCV, ECCV, AAAI, and TIFS, and holds over 30 patents. He has won multiple championships in competitions such as NeurIPS, CVPR, and ECCV DCASE. At Ant, he has overseen various business areas, including biometrics, content security, and identity security. His product achievements have been included in Gartner's Identity Authentication Market Guide and have won the IDC FinTech Real Results Award.

Report Title: Application and exploration of multimodal interaction in financial identity scenarios

Report Introduction:  

        This report introduces a series of practical breakthroughs in financial identity verification scenarios achieved through upgraded human-computer interaction. Ant's unified identity platform is Ant's digital identity infrastructure, providing reliable and convenient identity verification services to connect people, devices, and services from the physical world to the digital world. This report focuses on the team's exploration and implementation of financial identity verification in the AI 2.0 era through streaming video interaction, voice interaction, and sensor interaction. It provides a practical summary of industry experience at the intersection of human-computer interaction, financial technology, and identity security. Key topics include: 1. The challenge of deepfake attacks against traditional facial recognition and other identity verification methods, and new solutions offered through interaction. 2. Identity verification solutions for pan-intelligent devices such as smart glasses under the new human-computer interaction model of AI 2.0. 3. Opportunities and explorations in voice and multimodal interaction as a next-generation UI.

Introduction:
      Professor Wang Ye is a dual-appointed Assistant Professor at the Faculty of Science and Technology and the Faculty of Business Administration at the University of Macau. He received his PhD from ETH Zurich and his BA from Peking University. His research interests span information security, human-computer interaction, fintech, blockchain, and artificial intelligence. He has published over 40 papers in conferences and journals, including IEEE S&P, CHI, and WWW. He currently serves as an Executive Member of the CCF Blockchain Committee, an Executive Member of the CCF Computational Economics Committee, and Deputy Secretary-General of the CSIAM Blockchain Committee. He has also served for many years as Vice Chair and Program Committee member for conferences such as CCS, CHI, and WINE.

Report Title: Understanding user-perceived security risks and mitigation strategies in the web3 ecosystem

Report Introduction: The advent of Web3 technologies promises unprecedented levels of user control and autonomy. However, this decentralization shifts the burden of security onto the users, making it crucial to understand their security behaviors and perceptions. To address this, our study introduces a comprehensive framework that identifies four core components of user interaction within the Web3 ecosystem: blockchain infrastructures, Web3-based Decentralized Applications (DApps), online communities, and off-chain cryptocurrency platforms. We delve into the security concerns perceived by users in each of these components and analyze the mitigation strategies they employ, ranging from risk assessment and aversion to diversification and acceptance. We further discuss the landscape of both technical and human-induced security risks in the Web3 ecosystem, identify the unique security differences between Web2 and Web3, and highlight key challenges that render users vulnerable, to provide implications for security design in Web3. 

Introduction:
      Yi Xin is an associate professor at the Institute of Network Science and Cyberspace at Tsinghua University. His research focuses on application security and human-computer interaction. He has published over 40 papers at leading academic conferences and journals, including CHI, Ubicomp, and IJHCS, and has applied for and obtained over 20 invention patents. He has received numerous honors, including the ACM SIGCHI China Rising Star Award, the Young Talent Support Program in a Certain Field, the First Prize of the China Electronics Society's Science and Technology Progress Award, and the Outstanding Scientific and Technological Achievement Award from the China Artificial Intelligence Association.

Report Title: Interaction Security and Privacy in the Ubiquitous Computing Era

Report Introduction:

        In the era of ubiquitous computing, diverse human-computer interaction methods are deeply integrated into people's daily lives and work. With the development of artificial intelligence (AI), human-computer interaction can leverage multimodal, ubiquitous sensing capabilities to provide users with continuous, natural, and intelligent services. However, this multi-channel sensor data also exposes users' personal privacy to service providers and others, and the powerful reasoning capabilities of AI exacerbate these privacy risks. Furthermore, highly realistic and intelligent interactive interfaces have exerted unprecedented influence on users, raising a variety of security risks. This report will introduce security and privacy issues in human-computer interaction in the era of ubiquitous computing and explore methods for preventing and addressing them.

Introduction:
        Zhang Tengxiang, Alpha Power Co., Ltd. Dr. Zhang is a wearable product technology expert at the Labs and an executive member of the CCF Human-Computer Interaction and Ubiquitous Computing Committee. His research focuses on human-centered ubiquitous interaction technologies, including smart wearable devices, low-power sensing technologies, and mixed reality interaction systems. He has published over 20 papers in conference journals such as CHI and IMWUT and has been granted six patents. He has served as the project leader for projects and sub-projects under the National Key R&D Program and as the project leader for the National Natural Science Foundation of China's Young Scientists Fund. He has an interdisciplinary background in computer science, electronics, and biomedicine. He was formerly an associate researcher at the Institute of Computing Technology, Chinese Academy of Sciences, and has experience in semiconductors and consumer electronics.

Report Title: SparQi DevKit: Building an open data foundation for the next generation of wearable AI

Report Introduction:

        As the multimodal sensing capabilities of wearable devices continue to grow, open, standardized, and high-quality data collection platforms are becoming a critical foundation for AI research and industrial implementation. This report will introduce SparQi DevKit, an open wearable data platform for research institutions, medical institutions, and startups. Using multimodal wearable devices as the entry point, the platform will build a complete development data circulation ecosystem centered around data collection, sharing, model training, evaluation, and trading. Furthermore, SparQi DevKit introduces "data as a service" and "model matching" mechanisms to stimulate the value flow of wearable development data. We will share the platform's system design, data desensitization and trading mechanisms, collaborative models, and how to establish a truly "open and reusable" AI development infrastructure across hardware, data, and AI.