FB pixel

Qualcomm announces AI vision platform for IoT

 

Qualcomm has announced the Vision Intelligence Platform, which features a family of system-on-chips (SoCs) leveraging 10nm FinFET processing technology and designed for IoT robots, cameras, and smart home devices.

The QCS605 and QCS603 provide on-device image processing and machine learning, with energy and thermal efficiency, according to the company. They integrate Qualcomm’s most advanced image signal processer (ISP) and the company’s AI Engine, and include an ARM based multicore CPU, vector processing and GPU. The platform also includes SDKs and Qualcomm’s connectivity and security 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,” said Joseph Bousaba, vice president, product management, Qualcomm Technologies, Inc. “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. The platform is a premier launchpad for manufacturers and developers to create a new world of intelligent IoT devices.”

The Qualcomm Vision Intelligence Platform delivers up to 2.1 TOPS of compute performance for deep neural network inferences with the Qualcomm AI Engine and Snapdragon Neural Processing Engine software framework. The framework also provides a host of tools for developers and OEMs.

The platform supports up to 4K video resolution at 60fps, or 5.7K at 30fps, and multiple simultaneous streams at lower resolution, and Qualcomm says it integrates the most powerful camera processor ever created by the company. The platform also features a range of audio capabilities, including support for natural language processing and speech recognition.

KEDACOM and Ricoh THETA are planning to develop products based on the Qualcomm Vision Intelligence Platform, according to the announcement.

The SoCs are currently sampling, with multiple SKUs for a variety of requirements. Reference designs for QCS605-based VR 360 cameras from Qualcomm Technologies and Altek Corporation are available now, and reference designs for a QCS603-based industrial security camera are expected to be available in the second half of the year.

Article Topics

 |   |   | 

Latest Biometrics News

 

Imprivata CEO tells Biometric Update Podcast why identity must evolve faster

A lot of people will tell you how fast the tech industry moves. Fran Rosch, the CEO of Imprivata, has…

 

Passenger growth, AI fraud push digital travel credentials toward tipping point

Digital travel credentials (DTCs) are at a crucial moment in their adoption as the travel industry undergoes profound structural changes,…

 

Thales makes strong debut in NIST’s FRIF fingerprint biometrics benchmark

New entries to NIST’s benchmark for large-scale fingerprint biometric capture and comparison software from Thales and Innovatrics show significant gains…

 

CCIA entreats US Supreme Court to intervene in Texas app store age check law

In the present historical moment, it is borderline comical to see advocacy groups for the technology industry insist that age…

 

The US counter-cartel fight is becoming an identity intelligence war

The creation of the Joint Interagency Task Force-Counter Cartel (JIATF-CC) under the U.S. Northern Command (NORTHCOM) marks more than another…

 

Bangladesh positions digital ID and wallets as economic infrastructure

Bangladesh is advancing a “One Citizen-One ID-One Digital Wallet” strategy that aims to link identity, payments and government services through…

Comments

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Biometric Market Analysis and Buyer's Guides

Most Viewed This Week

Featured Company

Biometrics Insight, Opinion

Digital ID In-Depth

Biometrics White Papers

Biometrics Events