Create a Dataset

Set up a new dataset and choose your vision task type.

📍

Step 1 of 3: Create a Dataset

Part of the dataset preparation quickstart. Next: Upload images.

Create a dataset to store your images and annotations. You'll choose your vision task type and configure storage settings.

⏱️ Time: ~3 minutes


Create your dataset

  1. In your Datature Vi dashboard, click Dataset in the left sidebar
Datature Vi dashboard with Dataset tab highlighted
  1. Click Create Dataset
Create Dataset button

Choose your dataset type

Select the vision task you want to accomplish:

Dataset type selection screen

Best for: Object detection, defect detection, product identification

Locate and classify multiple objects within images using bounding boxes.

Learn more →

Click Next after selecting your type.


Choose your data type

Select Image or Video based on your data:

Data type selection
📘

Most users choose Image

Unless you're specifically working with video data, select Image.

Click Next to continue.


Configure settings

Enter your dataset details:

Dataset configuration screen
  • Dataset name — Choose a descriptive name
  • Dataset description — Briefly describe your dataset (optional)
  • Dataset localization — Select Multi-Region for best performance

Click Next to review.


Review and create

Verify your settings in the summary screen:

Dataset summary screen

Click Create Dataset to finish.

Dataset created!

Your dataset is ready. Next, upload your images.


What's next?

Add images to your dataset using drag-and-drop or the Vi SDK.