swoonyk/object-detection

Standalone Object Detection Service for iAssist

This service is a containerized Flask application that provides object detection using YOLOv8. It is designed to be deployed as a separate microservice.

Deployment to Hugging Face Spaces

Follow the same steps as you did for the TTS service, but for this new service.

Step 1: Create a New Space

1. 1.Go to your Hugging Face profile and click on "New Space".
2. 2.Space name: iassist-detection-service (or similar).
3. 3.License: MIT.
4. 4.Space SDK: "Docker".
5. 5.Hardware: "CPU basic" (from the free tier).
6. 6.Click "Create Space".

Step 2: Upload Your Files

1. 1.Go to the "Files" tab of your new space.
2. 2.Click "Add file" and then "Upload files".
3. 3.Upload the files for this service:
4. 4.app.py
5. 5.requirements.txt
6. 6.Dockerfile
7. 7.yolov8n.pt (from the root of the iassist-segf-v2 project)
8. 8.This README.md file.
9. 9.Commit the files to the main branch.

Step 3: Build and Deploy

Hugging Face will automatically build and deploy your service. This will take some time.

Step 4: Get Your Service URL

Once deployed, find its public URL. The detection API endpoint will be at:

https://<your-username>-<your-space-name>.hf.space/api/detect

You will need this URL for the final step.