Sign In

Communications of the ACM

ACM TechNews

Transparent, Reflective Objects Now Within Grasp of Robots


View as: Print Mobile App Share: Send by email Share on reddit Share on StumbleUpon Share on Hacker News Share on Tweeter Share on Facebook
The system can't pick up transparent or reflective objects as efficiently as opaque objects, but it is more successful than depth camera systems alone.

Carnegie Mellon University roboticists have developed a new technique for teaching robots to pick up transparent objects.

Credit: Carnegie Mellon University

Roboticists at Carnegie Mellon University (CMU) have developed a technique that uses a color camera to enable robots to pick up transparent and reflective objects.

The color camera system they developed can recognize shapes based on color.

The researchers also used images from depth cameras of opaque objects. paired with color images of the same objects, to train the system to mimic depth camera systems and infer shapes.

They said the system cannot grasp transparent or reflective objects as efficiently as opaque objects, but it works better than depth camera systems alone.

Said CMU's Thomas Weng, "Our system not only can pick up individual transparent and reflective objects, but it can also grasp such objects in cluttered piles."

From Carnegie Mellon University School of Computer Science
View Full Article

 

Abstracts Copyright © 2020 SmithBucklin, Washington, DC, USA


 

No entries found