Robots with 3D Vision
Robotics is rapidly evolving, integrating various algorithms into the system of robots. A new algorithm, has been developed by a group of researchers enabling robots to recognize three-dimensional objects at one glance.
Burchfiel and George Konidaris, professors of computer science at Brown University, presented their research in July 2017, showing that robots could identify 75% items out of 908 shown to them, from a single vantage point, whereas previously, robots were only able to recognize the items 50% of the time.
They created a computer vision algorithm which trained robots to learn their surrounding area by first sifting through 4,000 three-dimensional object database. These objects ranged across ten different classes, including bathtubs, chairs, desks, sofas, dressers, monitors, night stands, beds tables, and toilets.
Conventional algorithms, however, only train robots to recognize whole objects such as the entire chair or sofa. Old algorithms also train robots to recognize parts of a whole summing them together, focusing on the appearance of objects by identifying differences and similarities.
Scientists have made use of extra computing resources that understand if an item is right-side up, inferring its 3D shape, as parts of the object would be hidden.
This, however, gave rise to a problem in computer vision, as objects in real life usually overlap.
Scientists have solved this problem by embracing the most advanced form of artificial intelligence, using artificial neural networks. These give rise to deep-learning algorithms, which processes information the way human brains learn.
This deep learning analyzing all of the pixels in an image and predicts one simple output. When objects are partially hidden, robots have a limited view of it, the input is then less complex than the output, resulting in a three-dimensional representation.
You may be interested
Smart Nanotechnology Enlivens Wooden Ships and ArtifactsMohit Joshi - January 28, 2019
Scientists developed an innovative approach using ‘smart’ nanocomposites to conserve 16th-century British warship, the Mary Rose, and its artifacts. Shipwrecks are preserved in sediments under cold water,…
Researchers Discovered New Properties of Ferroelectric CrystalMohit Joshi - January 25, 2019
Researchers discovered new properties of ferroelectric crystal, according to a study published on July 10, 2018. This study was conducted by a team of researchers from China.…
Researchers Develop Novel Antibody to Knockdown of Protein in VertebratesMohit Joshi - January 24, 2019
Researchers from Technische Universität Dresden developed a synthetic antibody that improves functional analysis of proteins. A research led by Dr. Jörg Mansfeld of the Biotechnology Center of…