image-inpainting

Restore Images Seamlessly with Large Hole Inpainting

26 Apr 2025
Restore Images Seamlessly with Large Hole Inpainting

In the realm of image processing, one of the most challenging tasks is restoring images with large gaps or defects. The "Large Hole Image Inpainting" service offers developers a powerful Cognitive Action designed to tackle this issue head-on. Utilizing the advanced Mask-Aware Transformer (MAT) technology, this service excels in filling large voids in images, resulting in high-fidelity and diverse outputs that are both photo-realistic and visually appealing.

Imagine a scenario where an important historical photo has suffered damage, or a creative project requires seamless integration of new elements into existing visuals. This Cognitive Action allows for swift restoration and enhancement of images with minimal artifacts, making it an invaluable tool for graphic designers, photographers, and content creators alike.

Prerequisites

To get started with the Large Hole Image Inpainting service, you'll need an API key for accessing the Cognitive Actions API and a basic understanding of making API calls.

Execute Large Hole Image Inpainting

The Execute Large Hole Image Inpainting action is engineered to restore images with significant gaps or damage, leveraging the capabilities of the Mask-Aware Transformer. This action is particularly effective in producing visually coherent images that retain the original aesthetic quality.

Input Requirements

To utilize this action, you need to provide the following inputs:

  • Image: The main image you wish to inpaint, which must be a 512x512 pixel URI.
  • Mask (optional): A mask overlay indicating the area to be inpainted, also sized at 512x512 pixels. Black areas represent regions to be restored. If not provided, a random mask will be generated.
  • Seed: An integer to control randomness in the output. A value of -1 will select a random seed.
  • Model: Choose between "places" and "celeba" models, with "places" as the default.
  • Noise Mode: Options include "const", "random", or "none", with "const" as the default.
  • Truncation Psi: A floating-point number that adjusts the balance between image quality and diversity, where 1 indicates no truncation.

Expected Output

Upon executing this action, you can expect a URI pointing to the inpainted image, seamlessly filling in the specified gaps or areas of damage.

Use Cases for this specific action

  • Restoring Vintage Photos: Perfect for historians and archivists looking to preserve and restore old photographs that have suffered wear and tear.
  • Creative Design Projects: Graphic designers can enhance visual content by integrating new elements into existing images without visible seams.
  • Content Creation: Marketing professionals can use this action to create polished promotional materials by removing unwanted objects or filling in gaps in product images.

```python
import requests
import json

# Replace with your actual Cognitive Actions API key and endpoint
# Ensure your environment securely handles the API key
COGNITIVE_ACTIONS_API_KEY = "YOUR_COGNITIVE_ACTIONS_API_KEY"
# This endpoint URL is hypothetical and should be documented for users
COGNITIVE_ACTIONS_EXECUTE_URL = "https://api.cognitiveactions.com/actions/execute"

action_id = "e60351eb-3ba0-4c8a-8fbb-2f6ddd5177cd" # Action ID for: Execute Large Hole Image Inpainting

# Construct the exact input payload based on the action's requirements
# This example uses the predefined example_input for this action:
payload = {
  "mask": "https://replicate.delivery/mgxm/afd9384f-ac1e-4e4a-bf04-ed558383dda7/mask2.png",
  "seed": -1,
  "image": "https://replicate.delivery/mgxm/77aa654a-36e1-477a-8e05-2ea1724a6791/test2.jpg",
  "model": "places",
  "noiseMode": "const",
  "truncationPsi": 0.9
}

headers = {
    "Authorization": f"Bearer {COGNITIVE_ACTIONS_API_KEY}",
    "Content-Type": "application/json",
    # Add any other required headers for the Cognitive Actions API
}

# Prepare the request body for the hypothetical execution endpoint
request_body = {
    "action_id": action_id,
    "inputs": payload
}

print(f"--- Calling Cognitive Action: {action.name or action_id} ---")
print(f"Endpoint: {COGNITIVE_ACTIONS_EXECUTE_URL}")
print(f"Action ID: {action_id}")
print("Payload being sent:")
print(json.dumps(request_body, indent=2))
print("------------------------------------------------")

try:
    response = requests.post(
        COGNITIVE_ACTIONS_EXECUTE_URL,
        headers=headers,
        json=request_body
    )
    response.raise_for_status() # Raise an exception for bad status codes (4xx or 5xx)

    result = response.json()
    print("Action executed successfully. Result:")
    print(json.dumps(result, indent=2))

except requests.exceptions.RequestException as e:
    print(f"Error executing action {action_id}: {e}")
    if e.response is not None:
        print(f"Response status: {e.response.status_code}")
        try:
            print(f"Response body: {e.response.json()}")
        except json.JSONDecodeError:
            print(f"Response body (non-JSON): {e.response.text}")
    print("------------------------------------------------")



## Conclusion
The Large Hole Image Inpainting service offers an innovative solution for developers seeking to restore and enhance images with large areas of damage or voids. With its high-quality results and diverse applications, it stands out as a valuable asset for anyone involved in image processing. By leveraging this Cognitive Action, you can effortlessly transform damaged visuals into stunning, cohesive images that meet your creative needs. 

Consider integrating this powerful action into your projects to enhance your image restoration capabilities and elevate your workflow.
← Back to All Articles