OnQ Blog

Made for the IoT: Meet the Qualcomm Vision Intelligence Platform

Combining powerful AI, computer vision, and camera technology for smarter IoT devices at the edge

We’ve been hearing a lot about how the flourishing IoT ecosystem will change every aspect of our world. That change isn’t just about connecting billions of devices though. Intelligence is where the promise of the IoT will be realized, and in turn, will enable OEMs to develop innovative devices that will push the ecosystem forward.

Think of it this way: A home surveillance camera that notifies you every time your Great Dane jumps up on the sofa might be entertaining, but how useful is it, really? Now imagine that camera is positioned at your front door and able to differentiate between your child who’s been locked out and a burglar. In the case of the former, it could connect with the door’s smart lock to allow entry, and if the latter, sound an alarm that notifies the local police.

This level of home security is possible, but it won’t happen overnight. For that camera to be truly effective, it needs to be connected and intelligent enough to be able to process and analyze data in real time locally on the device, so it can recognize the things that matter and take immediate action. That’s where the just-announced Qualcomm Vision Intelligence Platform, which includes Qualcomm Technologies’ first family of system-on-chips (SoCs) for the development of smart devices for the IoT, comes in.

The new platform demonstrates Qualcomm Technologies’ commitment to the Internet of Things, the confidence in providing a solution that is intended to serve as a launch pad for OEMs, and the need for more capable hardware and software to support this growing industry. The platform, anchored by the Qualcomm QCS605 and Qualcomm QCS603 SoC, is imbued with our latest developments in camera and AI. The two 10nm SoCs include advanced camera processing software, machine learning and computer vision software development kits (SDKs), and proven connectivity and security technologies — all of which are designed explicitly for next-generation IoT devices.

Computing at the edge

The Vision Intelligence Platform combines cutting-edge processing and AI with superior imaging, which is key because so many IoT applications, of today and tomorrow, utilize visual information. This goes beyond security cameras to include everything from ovens to robotic vacuum cleaners and drones — any device that relies heavily on information that comes from a camera.

The processing of that information could be done in the cloud, but that takes resources and time. Because the camera itself has the intelligence, it can decide how to respond based on what it knows instead of waiting for video data to be sent to the cloud and analyzed. This is what’s known as edge computing — using smart devices to make decisions and take actions where the action is, at the edge of the internet. The advantages of edge computing include faster processing, local control, better security and privacy, and the use of less network bandwidth. Integrating this technology will also push the IoT ecosystem forward, as developers move away from the cloud and focus on the capabilities of the device.

Processing this data at the edge requires significant horsepower, and the Vision Intelligence Platform can deliver thanks to our signature heterogeneous computing, which includes a powerful 8-core, 64-bit Qualcomm Kryo 300 CPU, Qualcomm Adreno 615 GPU, Qualcomm Hexagon 685 Vector Processor, Qualcomm Spectra 270 ISP, as well as DSPs for sensors and audio. With these components, and support for up to 2x2 802.11ac Wi-Fi with MU-MIMO, Bluetooth 5.1, the Qualcomm 3D Audio Suite, Qualcomm Aqstic Audio technologies, and Qualcomm aptX audio, developers have the cutting-edge tools needed to build IoT applications that can see, hear, and connect to the world around them.

The Vision Intelligence Platform also integrates the Qualcomm AI Engine, which includes the hardware listed above and software, such as the Qualcomm Snapdragon Neural Processing Engine (NPE), that support on-device AI. The NPE boasts analysis, optimization, and debugging capabilities that help developers and OEMs port trained networks into the platform. The AI Engine is compatible with Tensorflow, Caffe and Caffe2 frameworks, Open Neural Network Exchange interchange format, Android Neural Networks API, and the Qualcomm Hexagon Neural Network library, so developers have the freedom to use their preferred framework.

The Qualcomm Vision Intelligence Platform builds on the best of Qualcomm Snapdragon, together with advanced Qualcomm Technologies’ software to create an IOT platform that allow our customers to develop the connected, intelligent devices of the future. What will you make with it?

The Qualcomm QCS603 and QCS605 SoCs are currently sampling. A QCS605-based VR/360 camera reference design from Altek, a leading original design manufacturer for cameras, is available today. A QCS603 based industrial security camera reference designs will be available later in 2018.

Qualcomm Vision Intelligence Platform, Qualcomm QCS605, Qualcomm QCS603, Qualcomm AI Engine, Qualcomm Adreno, Qualcomm Kryo, Qualcomm Hexagon, Qualcomm Neural Processing Engine, Qualcomm Spectra, Qualcomm 3D Audio Suite, Qualcomm Aqstic, Qualcomm aptX, and Qualcomm Snapdragon are products of Qualcomm Technologies, Inc. and/or its subsidiaries.

Opinions expressed in the content posted here are the personal opinions of the original authors, and do not necessarily reflect those of Qualcomm Incorporated or its subsidiaries ("Qualcomm"). Qualcomm products mentioned within this post are offered by Qualcomm Technologies, Inc. and/or its subsidiaries. The content is provided for informational purposes only and is not meant to be an endorsement or representation by Qualcomm or any other party. This site may also provide links or references to non-Qualcomm sites and resources. Qualcomm makes no representations, warranties, or other commitments whatsoever about any non-Qualcomm sites or third-party resources that may be referenced, accessible from, or linked to this site.