Create MS COCO style dataset

python json tensorflow annotations mscoco
user20112015 picture user20112015 · Aug 7, 2017 · Viewed 10.6k times · Source

How to create a MS COCO style dataset to use with TensorFlow? Does anyone have an experience with this? I have images, and annotations, as well as ground truth masks. I need to convert them to be compatible with MS COCO and any help is appreciated. I can't find any open source tool to create COCO style JSON annotations.

TensorFlow MS COCO reads JSON files which I'm not very experienced with.

Answer

jsbroks picture jsbroks · Jan 20, 2019

I'm working on a python library which has many useful classes and functions for doing this. It's called Image Semantics.

Here is an example of adding masks and exporting them in COCO format:

from imantics import Mask, Image, Category

image = Image.from_path('path/to/image.png')
mask = Mask(mask_array)
image.add(mask, category=Category("Category Name"))

# dict of coco
coco_json = image.export(style='coco')
# Saves to file
image.save('coco/annotation.json', style='coco')

Related questions

Why can't Python parse this JSON data?

I have this JSON in a file: { "maps": [ { "id": "blabla", "iscategorical": "0" }, { "id": "blabla", "iscategorical": "0" } ], "masks": [ "id": "valore" ], "om_points": "value", "parameters": [ "id": "valore" ] } I wrote this script to print all of the JSON data: import json from pprint import pprint …

Read More
How do I write JSON data to a file?

I have JSON data stored in the variable data. I want to write this to a text file for testing so I don't have to grab the data from the server each time. Currently, I am trying this: obj = open(…

Read More
How to prettyprint a JSON file?

I have a JSON file that is a mess that I want to prettyprint. What's the easiest way to do this in Python? I know PrettyPrint takes an "object", which I think can be a file, but I don't know …

Read More