Object Recognition

Robot arms are getting smarter and smarter all the time, and now a team from Georgia Tech has taught them how to make their own tools. These resourceful robots are first taught to match form and function, then let loose on a series of everyday objects to build what they need to complete a given job.

Mattel is getting its toys-to-life aspirations on the road with its Hot Wheels brand. The popular die-cast cars have been given a modern virtual makeover with MindRacers, a game that combines physical toy cars and game tokens with digital racing.​

​Microsoft’s AR headset, the HoloLens, is an exciting prospect but isn't quite ready for the public. But as it inches closer to a consumer release, the company has revealed that the next version of the hardware will have an AI coprocessor built in, to help the headset recognize real-world objects.

CCTV systems may help security personnel watch for handguns in places such as airports, but the task of manually scrutinizing every person on every screen is still a daunting one. It was with this in mind that researchers recently developed a system that automatically recognizes guns on video.

​The ancient Egyptian god Horus lost his eye in a fight, before having it restored. A symbol of restoration, it's the perfect name for a new wearable device for visually-impaired people, which describes a user’s surroundings to help them avoid obstacles, recognize faces and objects, and read text.

New York startup Robotbase is turning to Kickstarter to build a highly ambitious robot that would serve as a smart personal assistant for your home. It would be able to converse with you, act as a personal stylist, and even tell bedtime stories to your kids.

MIT neuroscientists compared the visual object recognition performance of primate brains to that of the latest deep neural networks, which try to mimic our visual system. They found that the best-performing network performed comparably to a macaque monkey brain.

Researchers from the University of Washington and the Allen Institute for Artificial Intelligence have created the first fully automated computer program that teaches itself everything there is to know about a visual concept by associating words with images.

A new object-detection software system from Fujitsu for use with on-board vehicle cameras could help enhance driver awareness by improving recognition of potential hazards, including pedestrians and moving vehicles.

Scientists at Brigham Young University (BYU) have developed an an algorithm that can accurately identify objects in images or videos and can learn to recognise new objects on its own.

The Cognitive Patterns system is being developed to help robots understand what they perceive through their cameras and other sensors.

A new computer program is able to guess what people are drawing, based on very rough sketches.