LOS ANGELES--(BUSINESS WIRE)--CloudSight, Inc. (https://cloudsight.ai), the leader in visual cognition and understanding, is announcing the issuance of a new patent to further advance its deep learning neural network captioning technology. This will further the image recognition API to aid in true computer understanding of digital media with human-like specificity.
The CloudSight API is designed to be accessible for developers and companies to integrate image recognition for a range of devices and platforms. With a growing database of over half a billion images consisting of high quality metadata, CloudSight’s proprietary technology leverages hybrid human and machine intelligence to provide real-time understanding of visual media including automated captioning, image classification, fine-grained object recognition, and scene understanding. Utilizing curated real-world data to train deep learning neural networks, CloudSight API continues to learn on its own and improve over time.
The company’s newest patent, U.S. Patent 9,830,522, is entitled “Image Processing Including Object Selection.” The patent covers systems for selecting specific objects within images. The selections may be made, for example using a finger or eye tracking. The selected objects may then be highlighted and displayed along with image tags characterizing contents of the image. The patent covers systems for tagging images, in which a particular object within an image is selected by a user prior to tagging. The tagging is optionally responsive to the particular object selected. As such, the user can chose to receive image tags associated with a particular object.
“Our dream has been to create a technology that is fully capable of understanding images,” said Brad Folkens, CEO and Co-Founder of CloudSight. “We are continuing to fine tune the CloudSight technology with the goal of replicating human-like cognition and accuracy. This patent is another step forwards for our AI technology.”
CloudSight offers companies the tools needed to access powerful image and video cognition. The company provides educational institutions free access to its API in order to foster research of deep learning neural networks and encourage training of developers to use visual cognition technology.
CloudSight is a Los Angeles-based tech company founded in 2012. With a growing database of over half a billion images consisting of high quality metadata, CloudSight utilizes real world data curated over the last five years to train deep learning neural networks to automatically caption images. The startup is backed by $8 million in funding, has over 30 patents pending worldwide, and counts many top online destinations and Fortune 500s as customers. Its mission is to bring visual cognition and understanding to artificial intelligence, giving AI the ability to understand what it sees.