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