Qualcomm is bringing deep learning to devices powered by its Snapdragon 820 processors, helping them to offer the likes of facial recognition and object tracking.
The chipset maker’s Technologies arm is launching a software development kit in the second half of this year. The Snapdragon Neural Processing Engine is powered by its Zeroth Machine Intelligence Platform.
Qualcomm said the SDK could allow device manufacturers to offer features including scene detection, object avoidance, gesture control and natural language processing.
The manufacturer is targeting the SDK at devices such as smartphones, security cameras, vehicles and drones. It said the kit could be used by the likes of the healthcare, automotive and security industries.
Gary Brotman, Director of Product Management, Qualcomm Technologies, said: “The demand for untethered, mobile learning-driven user experiences is increasing rapidly and Qualcomm Technologies’ customers have been clamouring for tools to help them realize their product visions.
“With the introduction of the new Snapdragon Neural Processing Engine SDK, we are making it possible for myriad sectors, including mobile, IoT and automotive to harnesses the power of Qualcomm Snapdragon 820 and make high-performance, power efficient on-device deep learning a reality.”
Frederick Soo, Chief Technology Officer of NAUTO, the autonomous vehicle technology company, added: “The…SDK helps us bring deep learning to our connected camera and smart cloud network, and that helps us present a full, contextual picture of the driving environment to our commercial automotive fleet and auto insurance customers.”
Qualcomm said it is the first system on chip provider to offer its customers a mobile specific deep learning toolkit.
Its Zeroth technology is already used in visual intelligence software and advanced malware protection products.