COCO is a large image dataset designed for object detection, segmentation, person keypoints detection, stuff segmentation, and caption generation. This package provides Matlab, Python, and Lua APIs that assists in loading, parsing, and visualizing the annotations in COCO. Please visit http://cocodataset.org/ for more information on COCO, including for the data, paper, and tutorials. The exact format of the annotations is also described on the COCO website. The Matlab and Python APIs are complete, the Lua API provides only basic functionality.

In addition to this API, please download both the COCO images and annotations in order to run the demos and use the API. Both are available on the project website.

Please download, unzip, and place the images in: coco/images/

Please download and place the annotations in: coco/annotations/

For substantially more details on the API please see http://cocodataset.org/#download.

To support Windows build and python3 we had to make minor changes to:

• PythonAPI/setup.py
• PythonAPI/pycocotools/coco.py

To install this package, use pip as follows:

(dlwin36coco) pferr@MSI d:\repos
$ pip install git+https://github.com/philferriere/cocoapi.git#subdirectory=PythonAPI
Collecting pycocotools from git+https://github.com/philferriere/cocoapi.git#subdirectory=PythonAPI
  Cloning https://github.com/philferriere/cocoapi.git to c:\users\pferr\appdata\local\temp\pip-build-6n1mxmto\pycocotools
Installing collected packages: pycocotools
  Running setup.py install for pycocotools ... done
Successfully installed pycocotools-2.0

On Windows, for the avove to work, you must have the Visual C++ 2015 build tools on your path. If you don't, make sure to install them from here:

Then, run visualcppbuildtools_full.exe and select default options:

If you clone/fork the original repo, you may hit the following error when running the demo notebooks:

It happens because there is a pycocotools directory in the same folder as the notebooks. The python code being imported then is the local implementation of pycocotools, not the one installed with pip. To fix this issue, the demo notebooks have been moved to their own directory.