Enhance Your Images with zf-kbot Cognitive Actions: Inpainting and Prompt Generation

24 Apr 2025
Enhance Your Images with zf-kbot Cognitive Actions: Inpainting and Prompt Generation

In the rapidly evolving world of image processing, the ability to manipulate and generate content is invaluable for developers. The zf-kbot/inpaint-and-guess-prompt spec introduces a powerful Cognitive Action designed to modify images intelligently. With this action, you can perform image inpainting or generate prompts based on specific masked areas of an image. This guide will walk you through the capabilities of this action and provide you with the necessary steps to integrate it into your applications.

Prerequisites

Before you begin using the Cognitive Actions, you will need to have an API key for the Cognitive Actions platform. You'll typically pass this API key in the headers of your requests to authenticate your usage.

Cognitive Actions Overview

Perform Image Inpaint or Generate Prompt

Description:
This action allows you to either modify specific areas of an image using inpainting techniques or generate descriptive prompts based on masked regions. It supports two operational modes: image inpainting and prompt generation.

Category: image-processing

Input: The following JSON schema outlines the required and optional fields for this action:

{
  "image": "string",
  "mask": "string",
  "seed": "integer",
  "steps": "integer",
  "prompt": "string",
  "scheduler": "string",
  "configValue": "number",
  "expansionSize": "integer",
  "negativeInput": "string",
  "borderStrength": "number",
  "colorIntensity": "number",
  "predictionType": "string",
  "samplingMethod": "string",
  "fineEdgeControl": "string",
  "inpaintIntensity": "number"
}
  • Required Fields:
    • image: A URI pointing to the input image to be processed.
    • mask: A URI pointing to the mask image, indicating areas for processing.
  • Optional Fields:
    • seed: Initializes the random number generator, defaulting to 0.
    • steps: Number of iterations for processing (default: 20, range: 1-50).
    • prompt: Text to guide the processing (default: empty string).
    • scheduler: The scheduling method (default: "karras").
    • configValue: Level of detail (default: 5).
    • expansionSize: How much the processing region grows (default: 1, range: 1-4).
    • negativeInput: Text specifying what to exclude (default: empty).
    • borderStrength: Strength of the edges (default: 0.55).
    • colorIntensity: Intensity of color adjustments (default: 0.55).
    • predictionType: Prediction method ("standard" or "guess", default: "standard").
    • samplingMethod: Method used for sampling (default: "euler_ancestral").
    • fineEdgeControl: Enables fine control over edges (default: "disable").
    • inpaintIntensity: Strength of inpainting effects (default: 1).

Example Input:

{
  "mask": "https://replicate.delivery/pbxt/MYJY7ENusmrLv2bdygK666ZsxI5xlFTjbIlTx3SEKSlM7oGe/45988b42-c522-4732-a122-cde32497caca-mask.jpg",
  "seed": 0,
  "image": "https://replicate.delivery/pbxt/MYJY7k5z525kc9Ichp76uuU6WX8CwXeQQ3cDiFp9QGLLbK3J/mom_1.jpg",
  "steps": 20,
  "prompt": "white hair",
  "scheduler": "karras",
  "configValue": 5,
  "expansionSize": 1,
  "negativeInput": "",
  "borderStrength": 0.55,
  "colorIntensity": 0.55,
  "predictionType": "standard",
  "samplingMethod": "euler_ancestral",
  "fineEdgeControl": "disable",
  "inpaintIntensity": 1
}

Output: The action typically returns a JSON object containing the result of the processing:

{
  "type": "standard",
  "image": "https://assets.cognitiveactions.com/invocations/379eec1e-dff7-4b41-b29c-f442d1ad1013/e90d7c5d-2ce9-4f41-a015-40db6fa02f93.png"
}

Conceptual Usage Example (Python): Here's how you might call this action using Python:

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 = "432295b0-96b1-4cd8-b61c-4a266a9a7473" # Action ID for Perform Image Inpaint or Generate Prompt

# Construct the input payload based on the action's requirements
payload = {
    "mask": "https://replicate.delivery/pbxt/MYJY7ENusmrLv2bdygK666ZsxI5xlFTjbIlTx3SEKSlM7oGe/45988b42-c522-4732-a122-cde32497caca-mask.jpg",
    "seed": 0,
    "image": "https://replicate.delivery/pbxt/MYJY7k5z525kc9Ichp76uuU6WX8CwXeQQ3cDiFp9QGLLbK3J/mom_1.jpg",
    "steps": 20,
    "prompt": "white hair",
    "scheduler": "karras",
    "configValue": 5,
    "expansionSize": 1,
    "negativeInput": "",
    "borderStrength": 0.55,
    "colorIntensity": 0.55,
    "predictionType": "standard",
    "samplingMethod": "euler_ancestral",
    "fineEdgeControl": "disable",
    "inpaintIntensity": 1
}

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 snippet, you see how to structure the input payload correctly, including the action ID and the required fields. Note that the endpoint URL and request structure are hypothetical and should be adapted to the actual implementation.

Conclusion

The zf-kbot/inpaint-and-guess-prompt Cognitive Action provides an exciting opportunity for developers to enhance their image processing applications. By utilizing this action, you can effectively modify images and generate creative prompts based on specific areas of interest. Consider experimenting with different configurations and inputs to discover the full capabilities of this powerful tool. Happy coding!

← Back to All Articles