Skip to content

A personal playground to try out the tensorflow2.x object detection API capabilities.

Notifications You must be signed in to change notification settings

krumbs/machine-learning-sandbox

Repository files navigation

Work-through of Tensorflow2.x Object Detection API

Tech stack

  • Anaconda Python38
  • Tensorflow 2.2
  • CUDA v10.1
  • CuDNN 7.6.5
  • PIP

Getting started

Follow this tutorial to install all necessary software to start using the Tensorflow2 Object Detecting API Tensorflow2 Object Detecting API Tutorial

Install additional necessary packages with:

pip install -r requirements.txt

1. Webcam Object Detection

  • related directory: webcam-object-detection/

This example will connect to the available webcam. Detect objects and save the detected objects in each frame plus the label to file.

Example webcam detected output

Usage

Each subproject needs to be run from its respected directory as its root. I.e. in your terminal from object_detection/webcam-object-detection/ run

python main.py

2. Mobile Video Object Detection

Custom implementation based on Mobile Video Object Detection with Temporally-Aware Feature Maps.

"...an online model for object detection in videos designed to run in real-time on low-powered mobile and embedded devices"

About

A personal playground to try out the tensorflow2.x object detection API capabilities.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published