BEGIN:VCALENDAR VERSION:2.0 PRODID:IEEE vTools.Events//EN CALSCALE:GREGORIAN BEGIN:VTIMEZONE TZID:Asia/Kolkata BEGIN:STANDARD DTSTART:19451014T230000 TZOFFSETFROM:+0630 TZOFFSETTO:+0530 TZNAME:IST END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20260204T183330Z UID:97ACA14A-304B-41D5-B12B-2BB824937770 DTSTART;TZID=Asia/Kolkata:20240604T213000 DTEND;TZID=Asia/Kolkata:20240604T223000 DESCRIPTION:Video coding is a fundamental and ubiquitous technology in the modern society. Generations of international video coding standards\, such as the widely-deployed H.264/AVC and H.265/HEVC and the latest H.266/VVC\ , provide essential means for enabling video conferencing\, video streamin g\, video sharing\, e-commerce\, entertainment\, and many more video appli cations. These existing standards all rely on the fundamental theory of si gnal processing and information theory to encode generic video efficiently with a favorable rate distortion behavior. In recent years\, rapid advanc ement in deep learning and artificial intelligence technology has allowed people to manipulate images and videos using deep generative models. Among these\, of particular interest to the field of video coding is the applic ation of deep generative models towards compressing talking-face video at ultra-low bit rates. By focusing on talking faces\, generative models can effectively learn the inherent structure about composition\, movement and posture of human faces and deliver promising results using very little ban dwidth resource. At ultra-low bit rates\, when even the latest video\n\nco ding standard H.266/VVC is apt to suffer from significant blocking artifac ts and blurriness beyond the point of recognition\, generative methods can maintain clear facial features and vivid expression in the reconstructed video. Further\, generative face video coding techniques are inherently ca pable of manipulating the reconstructed face and promise to deliver a more interactive experience. In this talk\, we start with a quick overview of traditional and deep learning-based video coding techniques. We then focus on face video coding with generative networks\, and present two schemes t hat send different deep information in the bitstream\, one sending compact temporal motion features and the other sending 3D facial semantics. We co mpare their compression efficiency and visual quality with that of the lat est H.266/VVC standard\, and showcase the power of deep generative models in preserving vivid facial images with little bandwidth resource. We also present visualization results to exhibit the capability of the 3D facial s emantics-based scheme in terms of interacting with the reconstructed face video and animating virtual faces.\n\nSpeaker(s): Dr. Yan Ye\, \n\nVirtua l: https://events.vtools.ieee.org/m/422747 LOCATION:Virtual: https://events.vtools.ieee.org/m/422747 ORGANIZER:ieee.sps.sb.iitkgp@gmail.com SEQUENCE:15 SUMMARY:IEEE SPS SBC Webinar: Face video compression with generative networ ks (By Dr. Yan Ye) URL;VALUE=URI:https://events.vtools.ieee.org/m/422747 X-ALT-DESC:Description: <br /><p dir="ltr">Video coding is a fundamental an d ubiquitous technology in the modern society. Generations of internationa l video coding standards\, such as the widely-deployed H.264/AVC and H.265 /HEVC and the latest H.266/VVC\, provide essential means for enabling vide o conferencing\, video streaming\, video sharing\, e-commerce\, entertainm ent\, and many more video applications. These existing standards all rely on the fundamental theory of signal processing and information theory to e ncode generic video efficiently with a favorable rate distortion behavior. In recent years\, rapid advancement in deep learning and artificial intel ligence technology has allowed people to manipulate images and videos usin g deep generative models. Among these\, of particular interest to the fiel d of video coding is the application of deep generative models towards com pressing talking-face video at ultra-low bit rates. By focusing on talking faces\, generative models can effectively learn the inherent structure ab out composition\, movement and posture of human faces and deliver promisin g results using very little bandwidth resource. At ultra-low bit rates\, w hen even the latest video</p>\n<p dir="ltr">coding standard H.266/VVC is a pt to suffer from significant blocking artifacts and blurriness beyond the point of recognition\, generative methods can maintain clear facial featu res and vivid expression in the reconstructed video. Further\, generative face video coding techniques are inherently capable of manipulating the re constructed face and promise to deliver a more interactive experience. In this talk\, we start with a quick overview of traditional and deep learnin g-based video coding techniques. We then focus on face video coding with g enerative networks\, and present two schemes that send different deep info rmation in the bitstream\, one sending compact temporal motion features an d the other sending 3D facial semantics. We compare their compression effi ciency and visual quality with that of the latest H.266/VVC standard\, and showcase the power of deep generative models in preserving vivid facial i mages with little bandwidth resource. We also present visualization result s to exhibit the capability of the 3D facial semantics-based scheme in ter ms of interacting with the reconstructed face video and animating virtual faces.</p>\n<p>&nbsp\;</p> END:VEVENT END:VCALENDAR