Qualcomm debuts IoT Vision Intelligence Platform, amid takeover rumours
Paul Jacobs, ousted as Qualcomm chairman last month.

Qualcomm debuts IoT Vision Intelligence Platform, amid takeover rumours

US chip giant Qualcomm Technologies has announced its new Vision Intelligence Platform, which features a new system-on-chip (SoC) family purpose-built for Internet of Things (IoT) applications.

The system integrates Qualcomm’s Artificial Intelligence Engine, along with a heterogeneous computing architecture that includes an ARM-based multicore CPU, vector processor, and GPU, said the company.

The aim is to help manufacturers and developers create “a new world of intelligent IoT devices”.

“Our goal is to make IoT devices significantly smarter as we help customers bring powerful on-device intelligence, camera processing, and security,” said Joseph Bousaba, VP product management, Qualcomm Technologies.

“AI is already enabling cameras with object detection, tracking, classification and facial recognition, robots that avoid obstacles autonomously, and action cameras that learn and generate a video summary of your latest adventure, but this is really just the beginning.

“The Qualcomm Vision Intelligence Platform is the culmination of years of advanced research and development that brings together breakthrough advancements in camera, on-device AI, and heterogeneous computing.”

What’s in the box?

The new platform is optimised to support a range of connected device and IoT applications, including smart security cameras, sports and wearable cameras, virtual reality (VR) 360- and 180-degree cameras, robotics, smart displays, and more.

According to Qualcomm, it supports up to 4K video resolution at 60 frames per second (fps), and features a dual 14-bit Qualcomm Spectra image signal processor that itself supports dual 16-megapixel camera sensors.

Meanwhile, the new QCS605 and QCS603 SoCs are designed to deliver fast, efficient, low-power computing for applications such as on-device camera processing and machine learning, said the company.

The chips are built using 10 nanometer Fin Field Effect Transistor (10nm FinFET) technology. Using the 10nm design, electrons have a smaller distance to travel across the gate within the chip, and so the system is able to run faster and with reduced battery drain. The 3D FinFET structure also minimises leakage to create a more efficient system.

The Vision Intelligence Platform also includes Qualcomm’s’ camera processing software, machine learning, and computer vision software development kits (SDKs).

Display maker KEDACOM, and Ricoh – for its 360-degree Theta camera – are among those planning to develop products based on the new platform.

Private investigations

However, the announcement has been overshadowed by news of a different kind: manoeuvres to take the company private, as Qualcomm emerges battle-weary from the recent fight with Broadcom.

Rival chipmaker Broadcom’s hostile takeover bid for the company was blocked by the White House last month, as Internet of Business reported. The move was an early signal of an escalation of trade hostilities between the US and China and its allies.

In the US, CNBC reports that former chairman Paul Jacobs, who was ousted from Qualcomm in March, is pushing ahead with a strategy to take the company private.

Quoting sources close to the rumoured plan, CNBC said that Jacobs is gathering potential investors, including strategic partners and sovereign wealth funds, to make a fully financed bid in the next two months, and put himself at the helm.

The report suggests that one of the investors to be approached is ARM Holdings, which is the property of Japan’s SoftBank, owner of the Aldebaran and Boston Dynamics robot brands. ARM’s technology is at the core of most processors used in smartphones and tablets, including Qualcomm’s new IoT SoCs.

ARM has denied that it has talked to Jacobs about a possible acquisition involving Qualcomm.

Internet of Business says

News of the new IoT chipset and vision platform provides further evidence that faster, more efficient chips are needed to run the IoT, along with edge intelligence and processing. And it is also evidence that the IoT doesn’t mean lower quality.

Meanwhile, rumours of this latest financing and/or takeover battle suggest that change is not only afoot, but somehow inevitable for the beleaguered chip giant.

However, it would be ironic if – after the political manoeuvres that kept Qualcomm from being snapped up by a Singapore-headquartered company in the largest tech deal in history –Qualcomm ended up being part-financed and part-owned by a Japanese robotics behemoth, which has been buying up US and European IP for years.

Especially since Broadcom is currently engaged in moves to become a fully US run corporation, in the wake of a $12 billion share repurchase scheme, announced yesterday.

Would you like chips with them apples?

Read more: MIT develops low-power high-speed chips for IoT security

Read more: NEWSBYTE: ARM launches scalable chips for IoT machine learning

Read more: SoftBank forecasts one trillion ARM IoT chips in next 20 years

Chris Middleton
Chris Middleton is former editor of Internet of Business, and now a key contributor to the title. He specialises in robotics, AI, the IoT, blockchain, and technology strategy. He is also former editor of Computing, Computer Business Review, and Professional Outsourcing, among others, and is a contributing editor to Diginomica, Computing, and Hack & Craft News. Over the years, he has also written for Computer Weekly, The Guardian, The Times, PC World, I-CIO, V3, The Inquirer, and Blockchain News, among many others. He is an acknowledged robotics expert who has appeared on BBC TV and radio, ITN, and Talk Radio, and is probably the only tech journalist in the UK to own a number of humanoid robots, which he hires out to events, exhibitions, universities, and schools. Chris has also chaired conferences on robotics, AI, IoT investment, digital marketing, blockchain, and space technologies, and has spoken at numerous other events.