Enhance Your Applications with Image Inpainting: A Guide to the swk23/bail Cognitive Actions

24 Apr 2025
Enhance Your Applications with Image Inpainting: A Guide to the swk23/bail Cognitive Actions

In the fast-evolving world of image processing, the swk23/bail Cognitive Actions provide developers with powerful tools to create and enhance images using advanced techniques. One of the standout features in this spec is the ability to generate images through inpainting, allowing you to creatively modify images based on detailed prompts. This article will explore how to effectively integrate and utilize the Generate Image with Image Inpainting action to elevate your applications' visual content.

Prerequisites

Before diving into the capabilities of the Generate Image with Image Inpainting action, ensure you have the following set up:

  • An API key for the Cognitive Actions platform.
  • Basic knowledge of making HTTP requests and handling JSON data.
  • Familiarity with Python programming for example implementations.

For authentication, you will typically pass your API key in the request headers to access the Cognitive Actions API securely.

Cognitive Actions Overview

Generate Image with Image Inpainting

The Generate Image with Image Inpainting action allows developers to create enhanced images by providing a detailed prompt and various configuration options. This action supports different models and parameters, enabling you to tailor the image generation process to your specific needs.

Input

The input for this action requires a JSON object structured according to the following schema:

  • Required Fields:
    • prompt: A descriptive string that guides the image generation.
  • Optional Fields:
    • mask: (string, uri) An image mask for inpainting mode.
    • image: (string, uri) An input image for image to image or inpainting mode.
    • model: (string) Choose between "dev" or "schnell" for different inference steps.
    • width: (integer) Width of the generated image (only if aspect_ratio is custom).
    • height: (integer) Height of the generated image (only if aspect_ratio is custom).
    • fastMode: (boolean) Enable faster predictions.
    • randomSeed: (integer) Seed for reproducible image generation.
    • outputCount: (integer) Number of outputs to generate (1 to 4).
    • outputFormat: (string) Format of the output images (webp, jpg, png).
    • guidanceScale: (number) Scale for the diffusion process.
    • loraIntensity: (number) Strength of the main LoRA application.
    • outputQuality: (integer) Quality of output images (0 to 100).
    • promptStrength: (number) Strength of the prompt when using img2img.
    • imageResolution: (string) Approximate number of megapixels for the image.
    • imageAspectRatio: (string) Aspect ratio for the generated image.
    • alternativeWeights: (string) Load LoRA weights from specific model repositories.
    • extraLoraIntensity: (number) Strength of the extra LoRA application.
    • inferenceStepCount: (integer) Number of denoising steps.
    • additionalLoraWeights: (string) Load additional LoRA weights from specific model repositories.
    • safetyCheckerDisabled: (boolean) Disable the safety checker for generated images.

Example Input:

{
  "model": "dev",
  "prompt": "\"A dimly lit office on Coruscant with large windows overlooking the city skyline, casting a soft glow into the room. Senator Mon Mothma, a poised woman with short auburn hair and sharp, intelligent eyes, sits at a circular table. She wears an elegant but simple white senatorial gown with subtle embroidery, reflecting her dignified presence. A datapad rests in front of her, and her fingers are lightly touching it as she contemplates, her expression wary and thoughtful. Shadows stretch across the room, emphasizing the secrecy of the meeting. The atmosphere is tense yet refined, the weight of political intrigue evident in her posture.\"",
  "fastMode": false,
  "outputCount": 1,
  "outputFormat": "jpg",
  "guidanceScale": 3,
  "loraIntensity": 1,
  "outputQuality": 80,
  "promptStrength": 0.8,
  "imageResolution": "1",
  "imageAspectRatio": "21:9",
  "extraLoraIntensity": 1,
  "inferenceStepCount": 28
}

Output

The action returns an array of URLs pointing to the generated images based on the provided prompt and parameters.

Example Output:

[
  "https://assets.cognitiveactions.com/invocations/f809a5a5-f7de-47f7-82e9-effcf7662f50/be369a80-c3b4-4cc6-8cfa-3a6a82888e2e.jpg"
]

Conceptual Usage Example (Python)

Here is a conceptual Python code snippet demonstrating how to invoke the Generate Image with Image Inpainting action:

import requests
import json

# Replace with your Cognitive Actions API key and endpoint
COGNITIVE_ACTIONS_API_KEY = "YOUR_COGNITIVE_ACTIONS_API_KEY"
COGNITIVE_ACTIONS_EXECUTE_URL = "https://api.cognitiveactions.com/actions/execute"  # Hypothetical endpoint

action_id = "adf93ce0-4a3d-43b5-9a4e-710d64c86238"  # Action ID for Generate Image with Image Inpainting

# Construct the input payload based on the action's requirements
payload = {
    "model": "dev",
    "prompt": "\"A dimly lit office on Coruscant with large windows overlooking the city skyline, casting a soft glow into the room. Senator Mon Mothma, a poised woman with short auburn hair and sharp, intelligent eyes, sits at a circular table. She wears an elegant but simple white senatorial gown with subtle embroidery, reflecting her dignified presence. A datapad rests in front of her, and her fingers are lightly touching it as she contemplates, her expression wary and thoughtful. Shadows stretch across the room, emphasizing the secrecy of the meeting. The atmosphere is tense yet refined, the weight of political intrigue evident in her posture.\"",
    "fastMode": False,
    "outputCount": 1,
    "outputFormat": "jpg",
    "guidanceScale": 3,
    "loraIntensity": 1,
    "outputQuality": 80,
    "promptStrength": 0.8,
    "imageResolution": "1",
    "imageAspectRatio": "21:9",
    "extraLoraIntensity": 1,
    "inferenceStepCount": 28
}

headers = {
    "Authorization": f"Bearer {COGNITIVE_ACTIONS_API_KEY}",
    "Content-Type": "application/json"
}

try:
    response = requests.post(
        COGNITIVE_ACTIONS_EXECUTE_URL,
        headers=headers,
        json={"action_id": action_id, "inputs": payload}  # Hypothetical structure
    )
    response.raise_for_status()  # Raise an exception for bad status codes (4xx or 5xx)

    result = response.json()
    print("Action executed successfully:")
    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: {e.response.text}")

In this code, you need to replace the placeholder for your API key and ensure the endpoint is correct. The provided payload structure matches the required input schema for the action.

Conclusion

The Generate Image with Image Inpainting action from the swk23/bail Cognitive Actions offers developers a robust tool for enhancing images through detailed prompts and customizable parameters. By leveraging this action, you can easily create visually appealing content for your applications. Explore the various options available, experiment with different prompt styles, and take your image generation capabilities to the next level! Happy coding!

← Back to All Articles