To Make Just-Noticeable Difference (JND) Computable toward Visual Intelligence
Thursday, October 31, 2024
Just-Noticeable Difference (JND) refers to the minimum change of signals, be they naturally captured or AI-generated, to be sensed by users (be they humans, machines, or both). The JND formulation can be part of the solution to tackle big data received with limited resources available, as the prerequisite for user-centric, and/or green systems, in terms of computation, bandwidth, storage space, energy/battery usage, device cost/size, and environmental constraints. In this talk, a systematic overview will be first presented on visual JND models, because majority of research in the community so far has been for images and video. Then, the related research and applications up to date are to be reviewed, from conventionally handcrafted approaches to data-driven ones, with different types of users. Finally, possible future opportunities for JND modelling are to be discussed, inclusive of visual signal reconstruction, image synthesis, graphic rendering, generative AI, extension toward comprehensive multimedia (to audio, smell and haptic/thermal signals), cross-modality efforts, and so on.
Post Talk Link: Click Here
Passcode: z0!!j9S*
Speaker/s
Weisi Lin is currently a President’s Chair Professor in College of Computing and Data Science, Nanyang Technological University (NTU), Singapore, where he also serves as the Associate Dean(Research). He is a Fellow of IEEE and IET. He has been awarded Highly Cited Researcher since 2019 by Clarivate Analytics, and elected for the Research Award 2023, College of Engineering, NTU. He has been a Distinguished Lecturer in both IEEE Circuits and Systems Society (2016-17) and Asia-Pacific Signal and Information Processing Association (2012-13). He has been an Associate Editor for IEEE Trans. Neural Networks Learn. Syst., IEEE Trans. Image Process., IEEE Trans. Circuits Syst. Video Technol., IEEE Trans. Multim., IEEE Sig. Process. Lett., Quality and User Experience, and J. Visual Commun. Image Represent. He has been a TP Chair for several international conferences and is a General Co-Chair for IEEE ICME 2025. He believes that good theory is practical and has delivered 10+ major systems for industrial deployment with the technology developed.
Related
Polygenic Score Modeling to Investigate Genotype-Phenotype Associations
Read More
Trustworthy Machine Learning: Transparency, Collaboration, and Evaluation
Read More