BEGIN:VCALENDAR
VERSION:2.0
PRODID:IEEE vTools.Events//EN
CALSCALE:GREGORIAN
BEGIN:VTIMEZONE
TZID:America/Los_Angeles
BEGIN:DAYLIGHT
DTSTART:20250309T030000
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3
TZNAME:PDT
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20241103T010000
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11
TZNAME:PST
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20250226T004802Z
UID:F78E220E-FC22-4EF1-BC94-61C8E0536292
DTSTART;TZID=America/Los_Angeles:20250225T150000
DTEND;TZID=America/Los_Angeles:20250225T160000
DESCRIPTION:The growing demand for immersive\, interactive experiences has 
 underscored the importance of 3D data in understanding our surroundings. T
 raditional methods for capturing 3D data are often complex and equipment-i
 ntensive. In contrast\, my research aims to utilize unconstrained videos\,
  such as those from augmented reality glasses\, to effortlessly capture sc
 enes and objects in their full 3D complexity. As a first step\, I will des
 cribe a method to incorporate Epipolar Geometry priors in multi-view Trans
 former models to enable identifying objects across extreme pose variations
 . Next\, I will discuss my work "Contrastive Lift" on 3D object segmentati
 on using 2D pre-trained foundation models\, following which I will talk ab
 out addressing the same problem using language.\n\nSpeaker(s): Yash Bhalga
 t\n\nAgenda: \n- Invited talk from Yash Bhalgat the final year PhD student
  at University of Oxford's Visual Geometry Group (VGG) supervised by Andre
 w Zisserman\, Andrea Vedaldi\, Joao Henriques and Iro Laina.\n- Q/A Sessio
 n\n\nVirtual: https://events.vtools.ieee.org/m/468355
LOCATION:Virtual: https://events.vtools.ieee.org/m/468355
ORGANIZER:upalmahbub@yahoo.com
SEQUENCE:28
SUMMARY:From Video to Virtual: Object-centric 3D scene understanding from v
 ideos
URL;VALUE=URI:https://events.vtools.ieee.org/m/468355
X-ALT-DESC:Description: <br /><p>The growing demand for immersive\, interac
 tive experiences has underscored the importance of 3D data in understandin
 g our surroundings. Traditional methods for capturing 3D data are often co
 mplex and equipment-intensive. In contrast\, my research aims to utilize u
 nconstrained videos\, such as those from augmented reality glasses\, to ef
 fortlessly capture scenes and objects in their full 3D complexity. As a fi
 rst step\, I will describe a method to incorporate Epipolar Geometry prior
 s in multi-view Transformer models to enable identifying objects across ex
 treme pose variations. Next\, I will discuss my work "<em>Contrastive Lift
 </em>" on 3D object segmentation using 2D pre-trained foundation models\, 
 following which I will talk about addressing the same problem using langua
 ge.</p><br /><br />Agenda: <br /><ul>\n<li>Invited talk from Yash Bhalgat 
 the final year PhD student at University of Oxford's Visual Geometry Group
  (VGG) supervised by Andrew Zisserman\, Andrea Vedaldi\, Joao Henriques an
 d Iro Laina.</li>\n<li>Q/A Session</li>\n</ul>\n<p>&nbsp\;</p>\n<p>&nbsp\;
 </p>
END:VEVENT
END:VCALENDAR