Qualcomm a mobile chip maker wants to bring deep learning into all kinds of devices and also expand the IOT environment. With this goal in mind, they created the Neural Processing Engine (NPE) for its Snapdragon-series mobile processors.
The NPE software development kit is now available to all via the Developer Network, marking the first public release of the SDK. This opens up a lot of opportunities for AI integration on a range of devices including mobile phones, in car platforms, etc. The purpose of the framework is to make possible UX implementations like style transfers and filters (basically what Snapchat and Facebook does with their mobile app cameras) with more accurate applications on user photos, as well as other functions better handled by deep learning algorithms like scene detection, facial recognition, object tracking and avoidance, as well as natural language processing. That is basically anything you would normally route to powerful cloud servers for an advanced process, but done locally on a device instead.
Facebook who gained early access to the NPE has already started seeing five-fold improvements in performance for its AR features on images and live video when using Qualcomm’s Adreno GPUs on Snapdragon SoCs. Qualcomm’s NPE works with the Snapdragon 600 and 800 series processor platforms and supports a range of common deep learning frameworks including Tensorflow and Caffe2.
This could be a major advancement for Qualcomm as more and more companies are investing on AI to improve reliability and reduce network dependence. This could help them maintain relevance for whatever comes after mobiles in terms of dominant tech trend in the upcoming years.