Upload Data
Uploading image data to Roboflow
Last updated
Was this helpful?
Uploading image data to Roboflow
Last updated
Was this helpful?
You can upload images and annotations for object detection, classification, instance segmentation, and semantic segmentation.
You can add data to your Roboflow account by:
Using the Web Interface. (Fastest for starting with ≤1000 images)
Using .
To upload data, first create a project if you do not have one already. When you first create a project, you will be asked to upload images:
If you already have a project, click "Upload" in the project sidebar to upload images.
You can upload larger datasets using the roboflow-python CLI.
Install the roboflow-python CLI by running pip install roboflow, then upload datasets using the roboflow import command.
Roboflow can accept the following types of data:
JPG, PNG, WEBP, and BMP images
MOV and mp4 videos
File size:
20MB maximum per image
Pixel Dimensions:
178,956,970 pixels max
This allows for example 16,400 × 10,900 pixels (16:9 aspect ratio)
If an image's content is the exact same, it will only counted towards your usage once.
You retain all ownership rights in any content, information, or materials You post, submit, publish, display, or transmit
To prevent issues from arising during training, we sanitize class names both at upload/import and export. At upload, we perform the following:
Trimming leading/trailing whitespace
All whitespace (including newlines & tabs) are converted to a space
Double spaces are removed
/.[]#~* characters are replaced with a dash (-)
|'" characters are removed
Public Plan: If you are on the Public plan unless explicitly specified and arranged by Roboflow, your datasets will be public on Roboflow Universe.
Paid Plans (including Enterprise): Unless otherwise specified, your data is private to your account.
For a detailed howto, refer to detailed instructions in the , or .
Data formatted in any for classification, object detection, segmentation, and keypoint detection.
is a repository of over 200,000 datasets covering object detection, classification, and segmentation tasks. You can use Universe to find data for use in training the first version of a model, or for use in fine-tuning a model.
For example, if you , the merged dataset and the two original datasets exist in your account. Therefore, there is no charge for the merge, because the images are duplicates.
We value our users' data ownership. As laid out in (specifically in section 23B), you retain ownership over all images and videos you upload to Roboflow.
