BEGIN:VCALENDAR VERSION:2.0 PRODID:IEEE vTools.Events//EN CALSCALE:GREGORIAN BEGIN:VTIMEZONE TZID:US/Eastern BEGIN:DAYLIGHT DTSTART:20230312T030000 TZOFFSETFROM:-0500 TZOFFSETTO:-0400 RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3 TZNAME:EDT END:DAYLIGHT BEGIN:STANDARD DTSTART:20231105T010000 TZOFFSETFROM:-0400 TZOFFSETTO:-0500 RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11 TZNAME:EST END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20230914T135436Z UID:4B9FE9B6-5085-4C70-ACEA-FA428A58DBC5 DTSTART;TZID=US/Eastern:20230913T120000 DTEND;TZID=US/Eastern:20230913T130000 DESCRIPTION:Neural networks and other advanced image processing algorithms excel in a wide variety of computer vision and imaging applications\, but their high performance also comes at a high computational cost and their s uccess is sometimes limited. In this talk\, we explore hybrid optical-digi tal strategies to computational imaging that outsource parts of the algori thm into the optical domain. Using such a co-design of optics and image pr ocessing\, we can design application-domain-specific cameras or compute pa rts of a convolutional neural network in optics. Optical computing happens at the speed of light and without any memory or power requirements\, ther eby opening new directions for intelligent imaging systems.\n\nCo-sponsore d by: Fairleigh Dickinson University\n\nSpeaker(s): Dr. Gordon Wetzstein \ , \n\nAgenda: \nNeural networks and other advanced image processing algori thms excel in a wide variety of computer vision and imaging applications\, but their high performance also comes at a high computational cost and th eir success is sometimes limited. In this talk\, we explore hybrid optical -digital strategies to computational imaging that outsource parts of the a lgorithm into the optical domain. Using such a co-design of optics and ima ge processing\, we can design application-domain-specific cameras or compu te parts of a convolutional neural network in optics. Optical computing ha ppens at the speed of light and without any memory or power requirements\, thereby opening new directions for intelligent imaging systems.\n\nVirtua l: https://events.vtools.ieee.org/m/367141 LOCATION:Virtual: https://events.vtools.ieee.org/m/367141 ORGANIZER:tan@fdu.edu SEQUENCE:38 SUMMARY:Next-generation Computational Imaging and Display Systems Engineeri ng based on End-to-end Optimization of Physical Structure and Signal Proce ssing URL;VALUE=URI:https://events.vtools.ieee.org/m/367141 X-ALT-DESC:Description: <br /><p>Neural networks and other advanced image p rocessing algorithms excel in a wide variety of computer vision and imagin g applications\, but their high performance also comes at a high computati onal cost and their success is sometimes limited. In this talk\, we explor e hybrid optical-digital strategies to computational imaging that outsourc e parts of the algorithm into the optical domain. Using such a co-design o f optics and image processing\, we can design application-domain-specific cameras or compute parts of a convolutional neural network in optics. Opti cal computing happens at the speed of light and without any memory or powe r requirements\, thereby opening new directions for intelligent imaging sy stems.</p><br /><br />Agenda: <br /><p>Neural networks and other advanced image processing algorithms excel in a wide variety of computer vision and imaging applications\, but their high performance also comes at a high co mputational cost and their success is sometimes limited. In this talk\, we explore hybrid optical-digital strategies to computational imaging that o utsource parts of the algorithm into the optical domain. Using such a co-d esign of optics and image processing\, we can design application-domain-sp ecific cameras or compute parts of a convolutional neural network in optic s. Optical computing happens at the speed of light and without any memory or power requirements\, thereby opening new directions for intelligent ima ging systems.</p> END:VEVENT END:VCALENDAR