There are a few places where you can upload data. For example, you can upload data to your account by clicking on the Upload Data button under the Data List page. You'll be taken to the page below.
You can also upload data to your account and simultaneously add them to a particular project through the Project Settings page.
By uploading data through the Project Settings page, you can both 1) upload them to your account and 2) assign the data to that project at the same time.
There are three ways to upload data to an account: 1) by File Upload, and 2) by CSV Upload, and 3) by Cloud Upload, 4) by SDK.
To upload by file, simply drag-and-drop files from your local computer, choose from the list of datasets on your account (or create a new dataset), and then hit Upload.
To upload by CSV, please first download a CSV template from the upload page. You'll see two columns in the CSV file -- a dataKey and URL. The dataKey field is the file name you would like to use for each raw data, and the URL field is the location where each raw data is stored on the web. Please fill these out accordingly and upload the CSV file.
Any image URLs which are publicly accessible and CORS enabled will work fine. To create public URLs for your own images in S3 Bucket, please refer to the documentation here :
Note: we don't store images uploaded through CSV Upload. We only store URL strings you have uploaded.
Select Destination Dataset, Select Data Type
Create a new dataset or select from the list of existing datasets.
Select the Data Type and the type of Cloud Storage.
Select Cloud Storage
Google Drive (Google Cloud Platform)
Naver Cloud (Coming Soon)
Select Integration Once you complete the setup process, select the type of Cloud Storage from the drop-down list. Select Data Storage Method
Superb AI Storage
Uploaded files are stored in our Superb AI’s cloud.
Your Own Storage
You give read-only access to your files. We do not create or store any copy of your data.
If you select this option, you may be charged for data access traffic.
i.e.) AWS S3 - Accessing 100,000 images of size 300KB, 20 times each, will incur about $70 of Amazon S3 pricing.
Select Folder Select the folder with the data you would like to upload. You will get a notification when the upload is complete.
Please refer to the SDK documentation here:
When uploading raw data, you can only upload files that match the project's data type. (i.e. you may not upload video files to a project configured for image data)
For video projects, each frame is considered as one data asset. If you have a video with a hundred frames, you will be metered for data storage of a hundred data assets.
Here is a list of supported file extensions:
.png, .jpg and .bmp
Sequence of Images: .png, .jpg, .bmp
Video: .mp4 (available for Cloud Upload only, limited to 100MB or 5 minutes per each file)
Once files are selected for upload, you may either choose from a list of existing datasets or create a new dataset before clicking Upload. Please note that datasets uploaded under Data List will not belong to any project in particular, and you must separately select them from the Data List and assign them to a particular project.
If the dataset has already been uploaded to your account and you want to assign it to a project, you can do that in the Data List page.
In order to upload a dataset and simultaneously assign to a project, please go to the project Settings and click on the Upload Data menu.
To download the project’s raw data, please follow the following steps
Go to Project Settings > General Tab > Raw Download.
Refreshbutton, which is located next to the Download button.
When it is ready, the
Download button will get activated.
To copy the download link, hit the
Any other questions? E-mail us at [email protected]