BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:From Pixels to 3D Motion: Modelling the Physical Natural World fro
 m Images - Elliott (Shangzhe) Wu
DTSTART:20250708T130000Z
DTEND:20250708T140000Z
UID:TALK233422@talks.cam.ac.uk
CONTACT:Miss Naidine Escoffery
DESCRIPTION:Pixel-based generative models nowadays excel at creating compe
 lling images\, but often struggle to preserve basic physical properties\, 
 such as shape\, motion\, material\, lighting\, etc. These are often critic
 al pieces bridging computer vision towards a wide range of real-world engi
 neering applications\, from interactive VR\, robotics\, design\, and manuf
 acturing\, to scientific domains like biology and medical analysis. One fu
 ndamental challenge in computer vision is shifting from modeling pixel dis
 tributions to modeling physics-grounded representations\, and characterizi
 ng 3D motion is a key stepping stone along this path.\n\nIn this talk\, I 
 will mainly discuss a line of research that attempts to model dynamic 3D o
 bjects from casually recorded\, in-the-wild images and videos\, without an
 y direct 3D supervision--an approach applicable to various natural objects
  such as wildlife. The resulting model can then turn a single image into a
 n animatable 3D asset in feed-forward fashion and generate 3D animations i
 nstantly.\n
LOCATION: Cambridge University Engineering Department\, CBL Seminar room B
 E4-38.  For directions see http://learning.eng.cam.ac.uk/Public/Directions
END:VEVENT
END:VCALENDAR