Can I Build a Real-Time Object Detection Space with Flask or FastAPI on Hugging Face?

Hello Hugging Face community,

I’m planning to create a Hugging Face Space for real-time object detection, using Flask or FastAPI as the backend to process images or video streams with models like YOLO or DETR from the Hugging Face Space.

I have two questions:

  1. Is it practical to run real-time object detection in a Space using Flask or FastAPI? What are the key limitations or best practices for deployment on Hugging Face Spaces?

  2. I’m worried about violating Hugging Face’s policies. Could this type of Space risk my account being flagged or blocked? What steps can I take to ensure compliance with Hugging Face’s Terms of Service?

Any advice, example Spaces, or links to relevant documentation would be greatly appreciated. Thank you!

Best,
Danh Tran (danhtran2mind).

1 Like

1

I think Gradio’s backend is FastAPI, so I think it should be possible…
I don’t know much about Flask.

2

I think 5. of this article mainly refers to prohibited acts in Spaces.

1 Like

Hey, do you like cats. I love dogs. Thanks for your support.

1 Like

This topic was automatically closed 12 hours after the last reply. New replies are no longer allowed.