Visual Intelligence Summit: Oct 22 in San Francisco Get your ticket

Convert Annotation Formats EASY (26 available Formats 😱)

Lukas Kelsey-FriedemannPublished Mar 16, 2022
1 min read
SUMMARY

Computer vision datasets come in dozens of annotation formats, and switching between them manually is tedious. Roboflow handles this in two steps: upload your dataset in whatever format you have, and export it in any of 26 supported formats. The platform auto-detects the incoming format, so no manual conversion is needed before you can use the data with a different training framework or tool.

Computer vision datasets come in a wide variety of formats, and sometimes they’re less than ideal. Roboflow solves this problem by offering 26 annotations formats for dataset uploading and downloading.

In two steps you can:
1) upload your dataset to Roboflow

and then,
2) download your dataset in virtually any annotation format.

Step-by-step video tutorial of converting datasets with Roboflow

1 ) Upload Your Dataset

After creating a new workspace, drag & drop or select the folder containing your dataset in the annotation page on your project's page

Roboflow will automatically detect the format, so you don't have to worry about anything.

You can then optionally modify your dataset by uploading more images, annotating them, and even combine datasets.

2) Download Your Dataset

Generate a new version of your dataset (or select a pre-existing version) and then click Export. You'll be prompted to choose from the 26 annotation formats available.

Export your dataset in 26 annotation formats with Roboflow

That's It!

You have 26 different annotation formats at your finger tips for free with Roboflow. Enjoy!

LatestFrom the Blog

Get started

Build on the Platform

For developers, engineers, and technical founders who want to get hands on. Try the free tier; the docs are open.

Bring it into your operation

For heads of AI, operations leaders, and enterprise teams. Bring a known problem, or work with us to find the first one worth taking on.