Elevate Your Image Generation with fszotyi/sdxl-car Cognitive Actions

22 Apr 2025
Elevate Your Image Generation with fszotyi/sdxl-car Cognitive Actions

In the realm of artificial intelligence, image generation has progressed remarkably, allowing developers to create stunning visuals with minimal effort. The fszotyi/sdxl-car spec introduces a powerful Cognitive Action designed to enhance your application's image generation capabilities. This action enables you to generate images using inpainting and img2img techniques, while offering customizable parameters for tailored outputs. In this article, we'll explore how you can leverage this Cognitive Action to create high-quality images effortlessly.

Prerequisites

Before you dive into using the Cognitive Actions, ensure you have the following:

  • An API key for accessing the Cognitive Actions platform.
  • Basic understanding of JSON and API requests.
  • Familiarity with Python for conceptual code examples.

Authentication typically involves passing the API key in the request headers to access the Cognitive Actions services.

Cognitive Actions Overview

Generate Image with Inpainting and Refinement

Description:
This action allows you to generate images through advanced techniques like inpainting and img2img. You can refine the output with various styles and customize it using parameters such as prompt strength, guidance scale, and more, ensuring that you achieve high-quality and tailored visual creations.

Category: Image Generation

Input

The input for this action is structured in a JSON format, requiring several fields that guide the image generation process:

  • mask (string, required): A URI for the input mask used in inpainting mode.
  • seed (integer, optional): A random seed for reproducibility.
  • image (string, required): A URI for the input image used in img2img or inpainting modes.
  • width (integer, optional): The width of the output image in pixels (default: 1024).
  • height (integer, optional): The height of the output image in pixels (default: 1024).
  • prompt (string, optional): A guiding text prompt for image generation (default: "An astronaut riding a rainbow unicorn").
  • loraScale (number, optional): Scale for LoRA models (default: 0.6).
  • outputCount (integer, optional): Number of images to generate (default: 1).
  • refineStyle (string, optional): Refinement style to apply (default: "no_refiner").
  • guidanceScale (number, optional): Guidance scale for classifier-free guidance (default: 7.5).
  • schedulerType (string, optional): Scheduler algorithm to use (default: "K_EULER").
  • applyWatermark (boolean, optional): Adds a watermark to the images (default: true).
  • negativePrompt (string, optional): Elements to avoid in the image.
  • promptStrength (number, optional): Influence of the prompt on generation (default: 0.8).
  • numInferenceSteps (integer, optional): Total number of denoising steps (default: 50).

Example Input:

{
  "width": 1024,
  "height": 1024,
  "prompt": "a photo of TOK car, in f1 race",
  "loraScale": 0.7,
  "outputCount": 1,
  "refineStyle": "expert_ensemble_refiner",
  "guidanceScale": 7.5,
  "schedulerType": "K_EULER",
  "applyWatermark": false,
  "negativePrompt": "",
  "promptStrength": 0.8,
  "refinementSteps": 0,
  "highNoiseFraction": 0.95,
  "numInferenceSteps": 50
}

Output

The action typically returns an array of generated images as URLs. Each execution can yield different outputs based on the input parameters.

Example Output:

[
  "https://assets.cognitiveactions.com/invocations/301784d3-95a3-4a20-8053-52d751ca3e2b/9c5a5013-3c3e-4e4c-8697-10b8f687724f.png"
]

Conceptual Usage Example (Python)

Here’s a conceptual Python snippet demonstrating how to invoke the action using the provided input structure:

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 = "39efad7d-0ead-45e8-a43a-8227a5aa0e13" # Action ID for Generate Image with Inpainting and Refinement

# Construct the input payload based on the action's requirements
payload = {
    "width": 1024,
    "height": 1024,
    "prompt": "a photo of TOK car, in f1 race",
    "loraScale": 0.7,
    "outputCount": 1,
    "refineStyle": "expert_ensemble_refiner",
    "guidanceScale": 7.5,
    "schedulerType": "K_EULER",
    "applyWatermark": False,
    "negativePrompt": "",
    "promptStrength": 0.8,
    "refinementSteps": 0,
    "highNoiseFraction": 0.95,
    "numInferenceSteps": 50
}

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 snippet, replace the placeholders with your actual API key and endpoint. The action ID and input JSON payload are structured according to the requirements of the "Generate Image with Inpainting and Refinement" action.

Conclusion

The fszotyi/sdxl-car Cognitive Action for image generation offers a robust solution for developers looking to integrate advanced image creation into their applications. With customizable parameters and the ability to refine outputs, you can create stunning visuals that meet your specific needs. Explore further by experimenting with different inputs and integrating this action into various use cases within your projects!

← Back to All Articles