Edge TPU / TF-Lite samples.

About

Coral Edge TPU / TensorFlow-Lite samples (Python/C++, Raspberry Pi/Windows/Linux).

• Object detection by PiCamera or Video Capture.
• Image classifilcation by PiCamera or Video Capture.
• Object camouflage by PiCamera.
• Semantic Segmentation using DeepLab v3.
• Image segmentation by PiCamera

Environment

• Coral Edge TPU USB Accelerator
• Raspberry Pi (3 B+ / 4) + PiCamera or UVC Camera
• x64 PC(Windows or Linux) + Video file or UVC Camera
• Python3

Installation

• OpenCV with OpenCV's extra modules(3.4.5 or higher)
• Edge TPU Python library (Get started with the USB Accelerator)

Usage

Image classification:
$ python3 ./object_detection_capture_picamera.py --model=<PATH_TO_edgetpu.tflite> --label=<PATH_TO_LABELS_TXT>

Full command-line options:
$ python3 ./xxxx.py -h

Reference

• Get started with the USB Accelerator
• TensorFlow models on the Edge TPU
• Models Built for the Edge TPU