BEGIN:VCALENDAR VERSION:2.0 PRODID:IEEE vTools.Events//EN CALSCALE:GREGORIAN BEGIN:VTIMEZONE TZID:America/New_York BEGIN:DAYLIGHT DTSTART:20250309T030000 TZOFFSETFROM:-0500 TZOFFSETTO:-0400 RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3 TZNAME:EDT END:DAYLIGHT BEGIN:STANDARD DTSTART:20241103T010000 TZOFFSETFROM:-0400 TZOFFSETTO:-0500 RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11 TZNAME:EST END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20250210T012533Z UID:5EA42F06-F890-4190-BA09-224B07EB8070 DTSTART;TZID=America/New_York:20250206T170000 DTEND;TZID=America/New_York:20250206T180000 DESCRIPTION:IEEE Circuits and Systems Society Montreal Chapter\, IEEE Solid State Circuits Society Montreal Chapter\, IEEE Montreal Section\, ReSMiQ – Regroupement Stratégique en Microsystèmes du Québec invite you to a ttend the talk by DR. YONG ZHAO.\n\n"With the rapid development of large l anguage models (LLM) featuring low cost and high performance\, they will s oon be applied to robots\, cars\, drones etc. The so-called embodied intel ligence technologies will be the next big market opportunity. However\, wh en these robots need to conduct certain tasks\, they require advanced 3D p erception ability of the environment. Although various 3D perception techn ologies\, such as LiDAR\, are also developing rapidly\, they\, similar to some other active perception devices\, have many problems and limitations. For example\, it may not produce effective reflections on mirror surfaces \, walls parallel to the line of sight\, etc. Moreover\, it only has a spa rse point cloud. Besides\, it lacks texture and color. Therefore\, for rob ots serving thousands of households\, binocular vision technology\, simila r to human vision\, is one of the key technologies for robot perception. B inocular vision algorithms have experienced a leapfrog development from tr aditional stereo matching algorithms based on image processing to a new ge neration of algorithms based on learning with deep nets. Although very hig h performance has been achieved\, the computational load is still relative ly high for embedded devices. Therefore\, further simplification and optim ization of the stereo matching algorithms is needed to enable widespread a pplication on low-cost mobile devices such as robots."\n\nSpeaker(s): Dr. Yong ZHAO\n\nRoom: 2.184\, Bldg: Concordia University - EV Building\, 1515 St. Catherine\, West\, Montreal\, Quebec\, Canada LOCATION:Room: 2.184\, Bldg: Concordia University - EV Building\, 1515 St. Catherine\, West\, Montreal\, Quebec\, Canada ORGANIZER:omair@ece.concordia.ca SEQUENCE:13 SUMMARY:Binocular Vision Stereo Matching: Development and Application of Al gorithms and Networks URL;VALUE=URI:https://events.vtools.ieee.org/m/466952 X-ALT-DESC:Description: <br /><p>IEEE Circuits and Systems Society Montreal Chapter\, IEEE Solid State Circuits Society Montreal Chapter\, IEEE Montr eal Section\, ReSMiQ &ndash\; Regroupement Strat&eacute\;gique en Microsys t&egrave\;mes du Qu&eacute\;bec invite you to attend the talk by DR. YONG ZHAO.&nbsp\;</p>\n<p>&nbsp\;</p>\n<p>"With the rapid development of large language models (LLM) featuring low cost and high performance\, they will soon be applied to robots\, cars\, drones etc. The so-called embodied inte lligence technologies will be the next big market opportunity. However\, w hen these robots need to conduct certain tasks\, they require advanced 3D perception ability of the environment. Although various 3D perception tech nologies\, such as LiDAR\, are also developing rapidly\, they\, similar to some other active perception devices\, have many problems and limitations . For example\, it may not produce effective reflections on mirror surface s\, walls parallel to the line of sight\, etc. Moreover\, it only has a sp arse point cloud. Besides\, it lacks texture and color. Therefore\, for ro bots serving thousands of households\, binocular vision technology\, simil ar to human vision\, is one of the key technologies for robot perception. Binocular vision algorithms have experienced a leapfrog development from t raditional stereo matching algorithms based on image processing to a new g eneration of algorithms based on learning with deep nets. Although very hi gh performance has been achieved\, the computational load is still relativ ely high for embedded devices. Therefore\, further simplification and opti mization of the stereo matching algorithms is needed to enable widespread application on low-cost mobile devices such as robots."</p> END:VEVENT END:VCALENDAR