Listen: Kostas Daniilidis on Geometric Methods in Computer Vision
Kostas Daniilidis, director of Penn Engineering Online and Ruth Yalom Stone Professor in Computer and Information Science, spoke to RobotsPodcast about his research in computer vision. Normally, capturing three-dimensional information from a visual scene requires specialized equipment: multiple cameras that can simultaneously view the subject from multiple angles, sensors that emit laser pulses and time how long it takes for them to return, or an array of markers on the subject itself that computers can track.
Now, Daniilidis and his lab members are working on new ways of gathering three-dimensional information from everyday two-dimensional pictures and video.
Read more at RoboHub.org