Smart glasses are an example of high-end output wearable devices; Whether they are used in industrial applications or to enhance our daily life experience, smart glasses are a model of computing vision applications, which need much stronger computing power than smart watches.

Design processor for smart glasses

If you are building an SOC for the most advanced wearable devices and need extreme performance per watt and per square millimeter, MIPs p-class CPU is an ideal choice. You can use TSMC’s latest 28nm HPC process node optimized for the Internet of things and let the processor run at low frequency to ensure that the design meets the strict power consumption and area requirements of wearable devices.

Because smart glasses need to see and understand their surrounding environment, SOC must have real-time visual processing function and pay attention to power consumption efficiency.

How to design chipsets for smart glasses and other high-end wearable devices

Example of chipset for smart glasses

Our powervr multimedia IP adds several unique features, making it have the following properties:

·Powervr Raptor ISP has the ability to transmit image histogram and video special statistical information to the encoder, which can improve performance and reduce power consumption; This feature also reduces the bit rate requirements of high-quality video coding

·Powervr GPU, Vpu and ISP have optimized data paths, which can reduce the memory delay of computational photography algorithm

·The dark color function (i.e. 10 bit yuv) is supported in all our powervr series (series6xt GPU, series 2 ISP, series 5 Vpu), which can improve image quality and coding efficiency

Powervr vision platform can run advanced computing vision algorithms

Smart glasses need to be able to communicate with the surrounding world. Our recently released ensigma combined RPU IP processor provides the latest high-speed wireless standards required for smart glasses: 802.11n/ac, classic Bluetooth and smart Bluetooth, digital and FM radio. If you need a detailed description of ensigma explore RPU functions, check this article and the printed release.

Case 1: Dhanush advanced from ineda

Dhanush advanced includes dual core, multithreaded MIPS class I CPU, which can easily execute multitasking applications when running Android or Linux operating system; In addition, Dhanush advanced includes a multimedia subsystem, which integrates powervr graphics and video encoding / decoding processing units.

Dhanush advanced can perform high-resolution rendering, complex image / video processing, and support low-power video recording and playback functions, so it is very suitable for high-end wearable devices.

Case 2: Sitara am437x from Texas Instruments

Texas Instruments (TI) recently announced the launch of a single core Sitara am437x processor equipped with powervr SGX GPU. The company also launched the Linux SDK and an evaluation suite to help developers start programming for high-end wearable and other Internet of things applications.

The first batch of projects applying Sitara SOC include ICIS, which is smart glasses from LaForge optical; You can find the usage evaluation of the device here.

ICIS smart glasses from LaForge optical

ICIS is a notification device connected through Bluetooth, which can provide real-time updates of social networks or navigation application information connected to GPS in smart phones or tablets.

ICIS provides an intuitive user interface

Leave a Reply

Your email address will not be published. Required fields are marked *