Hi there Mariana Logvinenko
Thanks for using QandA platform
I beleivee Vision does not natively support using your own Blob Storage (BYOS) for drag-and-drop scenarios, you can do this functionality by creating a custom workflow. Build a drag-and-drop interface using a front-end application that uploads images directly to your Blob Storage using Azure SDKs or REST APIs. Once the images are stored, you can trigger the Vision API to process them, using an Azure Function with a Blob Storage trigger.
Refer to the Azure Blob Storage SDK and Computer Vision API documentation for implementation details.
If this helps kindly accept the answer thanks much.