Apalis i.MX 8 Embedded Vision Starter Kit

Overview

Order your kit from Toradex here

The Apalis i.MX 8 Embedded Vision Starter Kit simplifies the development of computer vision and machine learning (ML) solutions.

Based on the i.MX 8QuadMax applications processor, the kit features connectivity, high reliability, reduced time-to-market, and ML optimizations in a compact form factor.

Amazon SageMaker Neo enables developers to train machine learning models once and run them anywhere in the cloud and at the edge.

AWS IoT Greengrass is used to manage secure communications with the cloud and enables the kit to deliver offline capabilities. A local user interface (UI) via an HDMI screen is included to provide live and historical data processing on a cloud dashboard.

Kit Contains:

Apalis i.MX 8QuadMax SoM
Ixora Carrier Board
Alvium 1500 series camera
12V DC 30W Power Supply
Lens
MIPI-CSI2 cable

Serial port adapter
USB to serial adapter
Apalis Heatsink Type 4
Alvium to Ixora adapter board
Null modem cable
Fan (with screws)

MoreLess

Apalis i.MX Embedded Vision Starter Kit

Apalis Demo Set Up

Apalis Demo UI Screen

Cloud-enabled Apalis AI Demo

With a few simple clicks, you can deploy a full end-to-end reference implementation of a device with an optimized neural network that can detect different kinds of pasta.

Specifications

Supported Devices

Processors and Microcontrollers

i.MX 8 Processors

i.MX8 : i.MX 8 Family – Arm^® Cortex^®-A53, Cortex-A72, Virtualization, Vision, 3D Graphics, 4K Video

MoreLess

Technical and Functional Specifications

Apalis iMX8 QuadMax System on Module	CPU: NXP^® i.MX 8QuadMax Applications Processor with to 2x Arm^® Cortex-A72, 4x Cortex-A53, 2x Cortex-M4F Armv8 64-Bit CPU Cores GPU: GC7000 3D RAM: 4GB LPDDR4 Flash memory: 16GB eMMC Operating Systems: Torizon, Embedded Linux, also available Yocto Project, Android and QNX On-board dual-band 802.11ac 2x2 MU-MIMO Wi-Fi and Bluetooth 5 Dimensions: 82.0 x 45.0 x 6.0 mm Temperature -40° to +85°C Industrial Grade System on Module design for products in demanding applications Pin-compatible with Apalis iMX6 and Apalis iMX8X
Ixora Carrier Board	Compatible with the full Apalis System on Module family HDMI and Dual Channel LVDS Audio In and Out 1x Gigabit Ethernet, 1x Mini PCIe, 1x mSATA 2x USB 3.0, 1x USB 2.0, 1x OTG 2x CAN, 2x I2C 1x SPI, 4x Analog inputs, Up to 40x GPIOs 1x Quad Lane MIPI CSI-2 Camera Interface Dimensions: 125 x 90mm Full Design freely available including complete Altium CAE Data
Alvium 1500 Series Camera	5MP MIPI CSI-2 Interface Compatible with a wide range of cameras Long-term availability Industrial Grade Image Signal Processor (ISP) on camera
Torizon Embedded Linux	Open Source Optimized for fast time-to-market Automotive Grade OTA option Built with Yocto Project Integration with Visual Studio and Visual Studio Code Docker Container Runtime
Amazon SageMaker Neo	Optimize Machine Learning models for Apalis iMX8 QuadMax Works with Apache MXNet, TensorFlow, PyTorch, XGBoost Bring your models to embedded devices

Design Resources

Get Started

Understanding your Apalis i.MX 8 Embedded Vision Starter Kit

AI at the Edge, Pasta Detection Demo with AWS

Get Started with the Apalis i.MX 8 Embedded Vision Starter Kit^{(REV
0
)}
- Y
- GET-STARTED-APALIS-IMX8X
- [ English ]

Documents

Data Sheet ( 1 )

Apalis i.MX 8 Data Sheet^{(REV
0
)}
- EXTERNAL
- nullkb
- 03 May 2020
- APALIS-IMX8-DATA-SHEET
- [ English ]

Users Guide ( 1 )

Get Started with the Apalis i.MX 8 Embedded Vision Starter Kit^{(REV
0
)}
- EXTERNAL
- nullkb
- 28 Apr 2020
- GET-STARTED-APALIS-IMX8X
- [ English ]

Get Help

Recommended Communities

i.MX Processors

Training and Events

Cloud-connected Computer Vision and Machine Learning Design

Type: On-Demand
Duration: 60:00

AI vision systems based on i.MX 8 are key elements for industrial applications requiring to recognize objects.

NOV18

Develop Effective Deep Neural Network Solution for the Edge Devices

Type: Webinar
Provider: NXP

Introduction about workflows and technologies for deep learning-based computer vision targeting edge devices.