STEREO VISION FROM A 10 CAMERA RING AND FPGA TO CREATE DEPTH IMAGES IN REAL TIME
Led by Marc Pollefeys, the research team at ETH Zurich have developed a 10 camera ring for TrimBot's body. From the ring, the robot is able to create stereo images. From stereo images, it is able to create depth images. And from those depth images, the robot is able to create a highly accurate 3D model of its environment.
With the use of FPGA hardware, the robot bypasses the need to send 10 images per second to a CPU, making image processing faster and 3D modelling possible in real time.
"You know the big question in the commununity is, how far can we go with just cameras?'
Prof. Marc Pollefeys, Department of Computer Science at ETH Zurich
and Director of Science at Microsoft Hololens
Hear Marc talk about the research at ETH Zurich in the 40sec video clip below, taken from a short documentary on the TrimBot2020 project, Cutting Hedge Research, which you can watch by clicking here.
For more information contact:
Corin Campbell: Corin.Campbell@ed.ac.uk
Prof. Robert Fisher, Consortium Coordinator: rbf@inf.ed.ac.uk
