New Visionati API promises developers more accurate tags and filters with facial recognition
Computer vision startup Visionati has announced the launch of a new API leveraging machine learning technologies with biometric facial recognition to apply what it says are more relevant tags and accurate filters than any other image or video recognition technology available, with a single API call.
The new API is an ideal tool for building applications or services with computer vision clarity comparable to human capacity, the company says, utilizing next-generation artificial neural networks and deep learning. Visual analysis features can be easily added to the API, according to the announcement, to help users and clients understand how they are interacting with the world. Pre-trained models provide frame-by-frame video analysis with enterprise level infrastructure Visionati says is highly scalable for high prediction accuracy.
The facial recognition technology integrated with API allows users to categorize, process and index a wide variety of data points and tags, including moods, themes, and emotions, according to Visionati.
Visionati also says the rich information its API extracts from real-world images and video provide higher confidence in object class detection accuracy, and it can quickly deliver actionable information and meaningful insights. Optical character recognition (OCR) capabilities translate text and hand-writing from images and enables searching.
“Your industry has already been revolutionized by computer vision; you may just not know it yet,” says Visionati Co-founder Josh Ellithorpe.
The company says its API makes image recognition easier and more accessible, and that it caters to both small-scale developers and large enterprises. It also says it is the first vision AI product on the market to accept Bitcoin Hash, Bitcoin, and Ethereum cryptocurrencies.
Several APIs have been launched recently to provide developers with facial recognition and emotion detection capabilities, including IDrive Face.