Table Of Contents
Table Of Contents

GluonCV: a Deep Learning Toolkit for Computer Vision

GluonCV provides implementations of state-of-the-art (SOTA) deep learning algorithms in computer vision. It aims to help engineers, researchers, and students quickly prototype products, validate new ideas and learn computer vision.

GluonCV features:

  1. training scripts that reproduce SOTA results reported in latest papers,
  2. a large set of pre-trained models,
  3. carefully designed APIs and easy to understand implementations,
  4. community support.

Supported Applications

Application Illustration Available Models
Image Classification:
recognize an object in
an image.
50+ models, including
ResNet, MobileNet,
DenseNet, VGG, ...
Object Detection:
detect multiple objects
with their bounding boxes
in an image.
Faster RCNN, SSD, Yolo-v3
Semantic Segmentation:
associate each pixel
of an image with
a categorical label.
FCN, PSP, DeepLab v3
Instance Segmentation:
associate each pixel of
an image with
an instance label.
Mask RCNN

Installation

Install MXNet

GluonCV depends on the recent version of MXNet. The easiest way to install MXNet is through pip. The following command installs CPU version of MXNet.

# the oldest stable version of mxnet required is 1.3.0
pip install mxnet>=1.3.0 --upgrade

# you can install nightly build of mxnet to access up-to-date features
pip install --pre --upgrade mxnet

Note

There are other pre-build MXNet binaries that enable GPU support and accelerate CPU performance, please refer to this tutorial for details.

Some training scripts are recommended to run on GPUs, if you don’t have a GPU machine at hands, you may consider to run on AWS.

Install GluonCV

The easiest way to install GluonCV is through pip.

pip install gluoncv --upgrade

# if you are eager to try new features, try nightly build instead

pip install gluoncv --pre --upgrade

Hint

Nightly build is updated daily around 12am UTC to match master progress.

Optionally, you can clone the GluonCV project and install it locally

git clone https://github.com/dmlc/gluon-cv
cd gluon-cv && python setup.py install --user