American multinational semiconductor and telecommunications equipment company, Qualcomm, gave birth to the Neural Processing Engine (NPE) for its Snapdragon-series mobile processors with an aim of enabling deep learning-based software development on all kinds of devices. And now, the company, which is considered as a world leader in 3G and next-generation mobile technologies, has made the NPE software development kit finally available to the world via its Qualcomm Developer Network.
The latest development marks the first public release of the SDK. According to tech experts, the kit can open a lot of great opportunities for Artificial Intelligence (AI) computing on a range of devices, right from mobile phones to in-car platforms and many more. This means, it could serve as a key to unlock a whole new world altogether.
By making the NPE kit available to public, Qualcomm hopes that humans will be able to achieve UX implementations like style transfers and filters with more accurate applications on user photographs, as well allow other functions like scene detection, facial recognition, object tracking and avoidance, as well as natural language processing to be better administered by deep learning algorithms. Basically, the kit would allow people to do anything that they normally do by taking the road of powerful cloud servers for advanced process, to be done locally right on their choice of device instead.
Qualcomm’s Neural Processing Engine functions smoothly with the Snapdragon 600 and 800 series processor platforms, and even boosts of supporting a range of common deep learning frameworks including the famous one like Caffe2 and Tensorflow.
Social networking giant Facebook is said to be one of the first ones to have landed its hand on Qualcomm’s NPE, and the Mark Zuckerberg led firm claims to be already registering a 5X improvements in the performance of its AR features on images and live video, when it is employing Qualcomm’s Adreno GPUs on Snapdragon SoCs. This gives us a glowing testimonial of Qualcomm’s NPE performance and quality.
Considering the fact that a lot of tech giants are now contemplating moving their AI-based computing functions from remote servers to local platforms so as to improve reliability and cut down on network connectivity requirements, Qualcomm can emerge as a big winner in this trend.