image-generation

Transform Your Images with Llkm's Inpainting Action

27 Apr 2025
Transform Your Images with Llkm's Inpainting Action

In the world of digital content creation, the ability to manipulate and generate images is invaluable. Llkm offers a powerful Cognitive Action that enables developers to generate images using advanced inpainting techniques. This action allows for both fast and high-quality outputs, making it an ideal solution for various applications. Whether you’re looking to restore an old photograph, create art, or enhance images for marketing, Llkm’s inpainting capabilities can significantly streamline your workflow.

This action provides flexibility with adjustable dimensions, aspect ratios, and output quality, alongside options for style customization using LoRA weights. Developers can leverage this technology to fulfill a wide range of use cases, from creative projects to practical applications in e-commerce and social media.

Prerequisites

To get started with Llkm, you will need a valid Cognitive Actions API key and a basic understanding of making API calls.

Generate Inpainting Image

The Generate Inpainting Image action is designed to create or modify images through inpainting techniques. This process allows you to fill in missing parts of an image or replace certain areas with new content based on a prompt. The action is particularly useful for graphic designers, marketers, and artists who require high-quality images tailored to specific needs.

Input Requirements: The action requires a structured input that includes:

  • Prompt: A textual description guiding the image generation, e.g., "a photo of LLKM - beautiful woman".
  • Image: An optional URI for an input image if you are performing image-to-image conversion.
  • Mask: A URI for the mask image in inpainting mode, which defines the areas to be altered.
  • Aspect Ratio, Dimensions, and Quality Settings: Customizable parameters such as width, height, output format, and quality level.

Expected Output: The output will be a URI link to the generated image, which can be used directly in various applications. For example, you might receive:

  • https://assets.cognitiveactions.com/invocations/d72b5018-11b1-46d3-b125-e6b378e9f646/54099787-0001-493d-937b-da7bd6820bd2.png

Use Cases for this specific action:

  1. Content Creation: Automatically generate unique images for blog posts, articles, or social media campaigns.
  2. E-commerce: Enhance product images by replacing backgrounds or adding elements to make them more appealing.
  3. Art and Design: Create artistic renditions or modify existing images to fit a specific theme or style.
  4. Restoration Projects: Fill in gaps in damaged photographs or artwork, revitalizing historical pieces.
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 = "7b374f93-0011-4ac7-af2f-69cf216c1c82" # Action ID for: Generate Inpainting Image

# Construct the exact input payload based on the action's requirements
# This example uses the predefined example_input for this action:
payload = {
  "prompt": "a photo of LLKM - beautiful woman ",
  "loraScale": 1,
  "numOutputs": 1,
  "outputFormat": "png",
  "outputQuality": 80,
  "extraLoraScale": 0.8,
  "inferenceModel": "dev",
  "imageAspectRatio": "4:5",
  "numInferenceSteps": 28,
  "diffusionGuidanceScale": 3.5
}

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

Llkm's Generate Inpainting Image action offers a robust solution for developers aiming to enhance their image generation capabilities. With its flexibility in inputs and outputs, alongside the ability to customize styles and quality, this action opens up numerous possibilities for creative and commercial applications. By integrating this action into your projects, you can significantly improve the efficiency and quality of your image-related tasks.

Explore how Llkm can transform your approach to image generation and consider the next steps in implementing this powerful tool in your applications.

← Back to All Articles