How I make machines see the 3D world | Daniel Cremers
TEDxTUM 2021 • Topics: 3D Technology, Artificial Intelligence, Computer Science, Computers, Future, Science, Technology

Clicking the video will play it from YouTube (transferring your IP address).

Teaching computers to understand the world from cameras has incredible potential for all kinds of applications: self-driving cars, medical image analysis, augmented reality, and many more. But the field of computer vision is full of unsolved problems. Daniel Cremers opens a window into the cutting-edge research aimed at solving some of these problems using techniques like convex optimization and deep neural networks.

About Daniel Cremers

Daniel Cremers is a physicist by training, a computer vision researcher by occupation, and head of the Chair of Computer Vision and Artificial Intelligence at the Technical University of Munich. His passion for computer vision and robotics has made him one of the most eminent researchers in the field as well as a pioneer in new algorithms for 3D computer vision. His work includes more accurate methods for camera-based reconstruction, more robust tracking approaches for moving cameras and novel ways of understanding 3D shapes. In 2016, he received the Gottfried Wilhelm Leibniz Prize for his pioneering research in image processing and pattern recognition.