Official English Documentation for ImageAI!

ImageAI is a python library built to empower developers, reseachers and students to build applications and systems with self-contained Deep Learning and Computer Vision capabilities using simple and few lines of code. This documentation is provided to provide detailed insight into all the classes and functions available in ImageAI, coupled with a number of code examples. ImageAI is a project developed by Moses Olafenwa and John Olafenwa , the DeepQuest AI team.

NOTE: ImageAI will switch to PyTorch backend starting from June, 2021

The Official GitHub Repository of ImageAI is


Installing ImageAI

ImageAI requires that you have Python 3.7.6 installed as well as some other Python libraries and frameworks. Before you install ImageAI, you must install the following dependencies.

or Tensorflow-GPU if you have a NVIDIA GPU with CUDA and cuDNN installed
pip install tensorflow-gpu==2.4.0
  • Other Dependencies

    pip install keras==2.4.3 numpy==1.19.3 pillow==7.0.0 scipy==1.4.1 h5py==2.10.0 matplotlib==3.3.2 opencv-python keras-resnet==0.2.0
  • ImageAI

    pip install imageai --upgrade

Once ImageAI is installed, you can start running very few lines of code to perform very powerful computer visions tasks as seen below.

Image Recognition

Recognize 1000 different objects in images

  • convertible : 52.459555864334106
  • sports_car : 37.61284649372101
  • pickup : 3.1751200556755066
  • car_wheel : 1.817505806684494
  • minivan : 1.7487050965428352

Visit Documentation

Image Object Detection

Detect 80 most common everyday objects in images.


Visit Documentation

Video Object Detection

Detect 80 most common everyday objects in videos.


Visit Documentation

Video Detection Analysis

Generate time based analysis of objects detected in videos.


Visit Documentation

Custom Image Recognition Training and Inference

Train new image new deep learning models on recognize custom objects


Visit Documentation

Custom Objects Detection Training and Inference

Train new YOLOv3 models to detect custom objects


Visit Documentation

Follow the links in the Content section below to see all the code samples and full documentation of available classes and functions.

Indices and tables