WiseEyeTM

View the repository on GitHub

 

 

 

Ultra low Power Endpoint AI Processors

Himax offers DRAM-less, ones-to-tens mill-watt AI processors to perform real time analytics in the endpoint itself.

The processors are designed to enable battery-powered intelligence vision, voice, and vibration applications.

 

 

 

WiseEye 2 AI Processor (WE2)

WE2 provides up to 50 GOPS with asymmetric multiprocessing architecture and DVFS power management.

 

Maximum ML computing

-       High performance 400MHz Cortex-M55

-       High efficiency 150MHz Cortex-M55

-       microNPU 400MHz Ethos-U55

-       Internal 2MB SRAMs and 512KB TCM

-       External 100MHz QSPI Flash, Max 16MB(128Mb)

Optimum energy efficiency

-       High-efficiency DC-DC

-       Internal power islands

-       DVFS power management

System-wide security

-       PUF-based hardware root-of-trust

-       PUF-based unique device ID

-       PUF-based TRNG

-       CryptoCell-312 crypto engine                                 

-       TrustZone security

-       Secure debug

Rich peripherals

-       MIPI CSI RX and TX

-       DVP and SDI

-       PDM and I2S

-       SD and SDIO

-       SPI, I2C, and I3C

-       UART, PWM and GPIO

 

 

Power Consumption Measurement - Power Modes

Note:

1. Measurement condition is in room temperature.

2. Facial landmark demo video showed in https://himaxwiseeyeplus.github.io/

3. Additional an on-board power switch needed to disable WE2 power source, except VDDIO_AON and VDDA_DCDC sources for GPIO wake up.

Chip Package WLCSP65 (2.3mm x 5.6mm)

 

Open-Source Unified Software Toolchain

An open-source unified toolchain for Ethos-U55 and Cortex-M55 simplifies developer creation of intelligence applications.

 

Optimized software flow for embedded and ML

-       TensorFlow Lite for Microcontrollers

-       Open-source Vela NN optimizer tool for Ethos-U55

-       Open-source CMSIS-NN library for Cortex-M55

-       Unified software flow for processor and microNPU integration

Unified software flow to accelerate ML performance

-       NN operators are accelerated by Ethos-U55 driver by default

-       Fallback to CMSIS-NN, then reference kernel by Cortex-M55

 

 

 

Himax WE2 board kit

Himax WE2 board kit information

-       Board kit online sales: https://www.seeedstudio.com/Grove-Vision-AI-Module-V2-p-5851.html

-       Board kit wiki: https://wiki.seeedstudio.com/grove_vision_ai_v2/

-       AI model assistant (Seeed’s SenseCraft ): https://wiki.seeedstudio.com/grove_vision_ai_v2_software_support/#-no-code-getting-started-with-sensecraft-ai-

-       Vision-based AI examples: Connect to open-source and web-based cloud: https://wiki.seeedstudio.com/connect_vision_ai_v2_to_ha/

-       Vision-based AI examples: Connect to phone-based app (Seeed’s SenseCap Mate App, iOS/Android): https://wiki.seeedstudio.com/connect_vision_ai_v2_to_ha/

Demo Videos

These demo videos are captured from WE2 silicon.

 

Facial landmark (12fps)

 

 

Human pose (7fps)

 

 

Face Mesh (7fps)

 

YoloV8n (9fps)

 

 

YoloV8n Pose(10fps)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Copyright Himax Technologies, Inc. All Rights Reserved.