Coral USB Accelerator

Featuring the Edge TPU, a small ASIC designed and built by Google, Coral USB Accelerator provides high performance Machine Learning (ML) inferencing with a low power cost over a USB 3.0 interface.

For example, it can execute state-of-the-art mobile vision models such as MobileNet v2 at almost 400 FPS, in a power-efficient manner.

Models are developed in TensorFlow Lite and then compiled to run on the USB Accelerator.

Features

  • Google Edge TPU ML accelerator coprocessor
  • USB 3.1 Gen 1 Type-C socket
  • Supports Debian Linux on host CPU
Anewtech-asus-google-coral-usb-accelerator

Anewtech-Coral-USB-Accelerator-edge-tpu

Edge TPU ML accelerator ASIC designed by Google that provides high performance ML inferencing for TensorFlow Lite models
Arm 32-bit Cortex-M0+ Microprocessor (MCU) T Up to 32 MHz max, 16 KB Flash memory with ECC, 2 KB RAM
Performs high-speed ML inferencing The on-board Edge TPU coprocessor is capable of performing 4 trillion operations (tera-operations) per second (TOPS), using 05 watts for each TOPS (2 TOPS per watt) For example, it can execute state-of-the-art mobile vision models such as MobileNet v2 at almost 400 FPS, in a power efficient manner See more performance benchmarks
Supports all major platforms Connects via USB to any system running Debian Linux (including Raspberry Pi), macOS, or Windows 10
Supports TensorFlow Lite No need to build models from the ground up TensorFlow Lite models can be compiled to run on the Edge TPU
Supports AutoML Vision Edge Easily build and deploy fast, high-accuracy custom image classification models to your device with AutoML Vision Edge
Connections USB 31 (gen 1) port and cable (SuperSpeed, 5Gb/s transfer speed)
Solutions for on-device intelligence
Anewtech-ai-coral-Object-detection

Object detection

Draw a square around the location of various recognized objects in an image.

Anewtech-ai-coral-Pose-estimation

Pose estimation

Estimate the poses of people in an image by identifying various body joints.

Anewtech-ai-coral-Image-segmentation

Image segmentation

Identify various objects in an image and their location on a pixel-by-pixel basis.