Keynotes
Title: DiffRender-VLA: Bridging 3D and 2D Robot Policies
Bio:
Professor Xiangjian (Sean) is currently the Chair Professor of Computer Science and Technology, Deputy Head of Computer Science School and the Director of Computer Vision and Intelligent Perception Laboratory at the University of Nottingham Ningbo China (UNNC). He is a National High-Level Talent of China, and in list of the ‘World Top 2% Scientists’ jointly released by Stanford University and Elsevier in the single-year database of 2022-2025 and the career-long database. He was the Professor of Computer Science and the Leader of Computer Vision and Pattern Recognition Laboratory at the Global Big Data Technologies Centre (GBDTC) at the University of Technology Sydney (UTS) from 2011-2022. He was involved in a team receiving a UTS Chancellor’s Award for Research Excellence through Collaboration in 2018. He led the UTS and Hong Kong Polytechnic University (PolyU) joint research project teams winning the champion for the 2019 VIP Cup, awarded by IEEE Signal Processing Society. He recently led a UNNC team achieving the 1st Place Overall in the Foundation Model Challenge for Ultrasound Image Analysis awarded by IEEE ISBI 2026. He has been carrying out research mainly in the areas of computer vision, data analytics and machine learning in the previous years. He has played various chair roles in many international conferences such as ACM MM, ICDAR, IEEE BigDataSE, IEEE IEEE TrustCom, IEEE CIT, IEEE AVSS, IEEE ICPR and IEEE ICARCV. He has received various competitive national or regional grants including five grants awarded by Australian Research Council (ARC) as a Chief Investigator. He has many high quality publications including ESI highly cited papers, papers in prestigious journals such as npj Digital Medicine (a Nature partner journal), Journal of the Association for Information Science and Technology and ACM Computing Surveys, IEEE Transactions journals such as IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), IEEE Transactions on Multimedia, etc., and in Elsevier’s journals such as Pattern Recognition, Knowledge-Based Systems, Expert Systems with Applications, Signal Processing, Information Fusion, Information Sciences, etc. He has also had papers published in premier international conferences and workshops such as AAAI, ACL, IJCAI, CVPR, ICCV, ECCV, ACM MM, ICDAR, WACV, etc. He is currently an Associate Editor of IEEE Transactions on Multimedia, Springer-Nature Computer Science Journal, the journal of Human-centric Computing and Information Sciences, etc.
Abstract
Robotic manipulation requires both spatial reasoning and visual perception, yet current Vision-Language-Action models (VLAs) excel at only one. 3D VLAs reason precisely but lack visual intuition, while 2D VLAs perceive richly but miss explicit spatial grounding. We introduce DiffRender-VLA, a differentiable rendering framework that bridges this gap. It generates images that embed 3D spatial semantics by localizing target actions, encoding geometry with color-coded directional beams, and learning adaptive viewpoints. This creates a closed gradient loop, allowing 2D VLA losses to directly improve 3D spatial representations. Experiments show state-of-the-art results, with an average +12.1% improvement in simulation and +17.5% on real robots, demonstrating effective transfer of geometric understanding into interpretable visual policies.
Technology for Health and Wellbeing in the Workplace
Dr. Mary Czerwinski is a Partner Research Manager of the Human Understanding and Empathy (HUE) Research Group at Microsoft Research. Mary’s latest research focuses primarily on behavior change and intervention design, health and wellness for individuals and productivity at work. Her research background is in visual attention and multitasking. She holds a Ph.D. in Cognitive Psychology from Indiana University in Bloomington. Mary received the ACM SIGCHI Lifetime Service Award, was inducted into the CHI Academy and received the Distinguished Alumni award from Indiana University’s College of Arts and Sciences. Mary is a Fellow of the ACM and the American Psychological Science Association. In 2022, Mary was inducted into the National Academy of Engineering. More information about Dr. Czerwinski can be found at her website: https://www.microsoft.com/en-us/research/people/marycz/.
Abstract
How can we create technologies to help us reflect on and potentially change our behavior, as well as improve our health and overall wellbeing both at work and at home? In this talk, I will briefly describe the last several years of work our research team has been doing in this area. We have developed wearable technology to help families manage tense situations with their children, mobile phone-based applications for handling stress and depression, as well as automatic stress sensing systems plus interventions to help users just in time. The overarching goal in all of this research is to develop intelligent systems that work with and adapt to the user so that they can maximize their personal health goals and improve their wellbeing.
Professor and Dean, Graduate School of Information Sciences, Tohoku University, Japan
