Documentation for interacting with the RoomAligner API, which is hosted at https://ashad001-roomaligner.hf.space. This API helps generate room structure suggestions based on uploaded images or URLs of room plans.

For embedding the API in an application or webpage, you can use an iframe as follows:

<iframe
  src="https://ashad001-roomaligner.hf.space"
  frameborder="0"
  width="850"
  height="450"
></iframe>

• Endpoint: /upload-room-image/
• Method: POST
• Description: This endpoint accepts an image of a room plan, processes it to classify objects, generates a natural language description, and provides suggestions for room structure improvement.

• URL: /upload-room-image/
• Method: POST
• Content-Type: multipart/form-data
• Parameters:
  • file: The room plan image file (JPEG, PNG, or BMP format).

curl -X POST "https://ashad001-roomaligner.hf.space/upload-room-image/" \
  -H "accept: application/json" \
  -H "Content-Type: multipart/form-data" \
  -F "file=@/path/to/your/image.jpg"

If the request is successful, the API returns a JSON response with:

• Natural language description of the room plan.
• The current room structure.
• Formatted room structure for better flow and space utilization.

{
  "natural_language_description": "The room contains a sofa, coffee table, and TV. The layout is suitable for a small living room.",
  "room_structure": "Current room layout includes the following furniture: Sofa (center), TV (front), Coffee Table (middle).",
  "formatted_room_structure": "We suggest moving the coffee table closer to the seating area and positioning the TV on the right wall to improve flow."
}

• 500 Internal Server Error: If something goes wrong while processing the image or generating suggestions, the API will return an error message in JSON format.

{
  "detail": "An error occurred: <error details>"
}

• Endpoint: /upload-room-image-url/
• Method: POST
• Description: This endpoint accepts the URL of an image of a room plan, processes it, and provides suggestions for room structure improvement.

• URL: /upload-room-image-url/
• Method: POST
• Content-Type: application/x-www-form-urlencoded
• Parameters:
  • image_url: The URL of the room plan image (JPEG, PNG, or BMP format).

curl -X POST "https://ashad001-roomaligner.hf.space/upload-room-image-url/" \
  -H "accept: application/json" \
  -d "image_url=https://example.com/path/to/room_image.jpg"

Similar to the file upload endpoint, the API returns a JSON response with:

• Natural language description of the room plan.
• The current room structure.
• Formatted room structure for better flow and space utilization.

To interact with the API, follow these steps:

1. File Upload:

   • Use an HTML <input> element with type="file" to allow users to upload the image of the room.

   <input type="file" id="roomImage" accept="image/jpeg,image/png,image/bmp">

2. API Request:

   • Use JavaScript's FormData to package the uploaded image.
   • Send a POST request to the /upload-room-image/ endpoint using fetch() or any preferred HTTP client (e.g., Axios).

   Example Using JavaScript's Fetch API:

   const uploadImage = async () => {
       const fileInput = document.getElementById('roomImage');
       const formData = new FormData();
       formData.append('file', fileInput.files[0]);
   
       try {
           const response = await fetch('https://ashad001-roomaligner.hf.space/upload-room-image/', {
               method: 'POST',
               body: formData
           });
   
           const result = await response.json();
           console.log('Natural Language Description:', result.natural_language_description);
           console.log('Current Room Structure:', result.room_structure);
           console.log('Formatted Room Structure:', result.formatted_room_structure);
       } catch (error) {
           console.error('Error uploading image:', error);
       }
   };

3. URL Upload:

   • You can also let users input an image URL and send it to the /upload-room-image-url/ endpoint.

   Example Using JavaScript's Fetch API:

   const uploadImageUrl = async () => {
       const imageUrl = document.getElementById('imageUrl').value;
   
       try {
           const response = await fetch('https://ashad001-roomaligner.hf.space/upload-room-image-url/', {
               method: 'POST',
               headers: {
                   'Content-Type': 'application/x-www-form-urlencoded',
               },
               body: `image_url=${encodeURIComponent(imageUrl)}`
           });
   
           const result = await response.json();
           console.log('Natural Language Description:', result.natural_language_description);
           console.log('Current Room Structure:', result.room_structure);
           console.log('Formatted Room Structure:', result.formatted_room_structure);
       } catch (error) {
           console.error('Error uploading image URL:', error);
       }
   };

4. Handle the Response:

   • Parse the JSON response and display the results on your webpage. You can show the natural language description, current room structure, and suggestions for improving the room layout.

CORS is enabled on the API, allowing requests from any origin (*). Therefore, frontend applications can make cross-origin requests directly without additional configuration.