Object recognition for robots
Robots’ maps of their environments can make existing object-recognition algorithms more accurate.
Robots’ maps of their environments can make existing object-recognition algorithms more accurate.
CSAIL team just misses winning the grand prize after programming a 400-lb humanoid robot to lift beams, climb stairs, and drive a car.
“Visual microphone” technology could lead to noninvasive identification of objects’ structural defects.
System designed to label visual scenes according to type turns out to detect particular objects, too.
New algorithm could enable household robots to better identify objects in cluttered environments.
The Association for Computer Machinery cites Devadas, Grimson, Morris, Rubinfeld, and Rus as having "provided key knowledge" to computing.
Deep-learning algorithm can weigh up a neighborhood better than humans.
Equipped with a novel optical sensor, a robot grasps a USB plug and inserts it into a USB port.
Algorithm tested aboard the International Space Station analyzes the rotation of objects in space.
Mint Solutions tackles medication errors with scanning system that ensures patients get the right pills.
Algorithm recovers speech from the vibrations of a potato-chip bag filmed through soundproof glass.
Techniques from natural-language processing enable computers to efficiently search video for actions.
New algorithm uses subtle changes to make a face more memorable without changing a person’s overall appearance.
CSAIL researchers are using a computational model that better understands peripheral vision to test the usability of MBTA subway maps.