Transform Images Using Pose Detection with jagilley/controlnet-pose Actions

24 Apr 2025
Transform Images Using Pose Detection with jagilley/controlnet-pose Actions

In the world of image processing, enhancing and transforming images containing human figures is a common challenge. The jagilley/controlnet-pose API provides a powerful Cognitive Action that leverages pose detection to facilitate this task. By integrating pose maps and text inputs, developers can create precise and condition-based image generation, making it an invaluable tool for various applications.

Prerequisites

To get started with the Cognitive Actions in this spec, you'll need an API key for the Cognitive Actions platform. This key will be used for authentication and must be included in the headers of your API requests. Ensure your development environment is set up to make HTTP requests, typically with libraries such as requests in Python.

Cognitive Actions Overview

Modify Images Using Pose Detection

This action enhances and transforms images by utilizing pose detection through the ControlNet model. It adapts Stable Diffusion to incorporate pose maps and text inputs, enabling precise and creative image generation.

Category: image-generation

Input

The following fields are required for invoking this action:

  • image (string, required): URL of the input image used in the process. Must be a valid URI.
  • prompt (string, required): Main text input that guides the model's output.

The action also includes several optional fields:

  • scale (number): The intensity of classifier-free guidance. Default is 9, with a range from 0.1 to 30.
  • denoisingSteps (integer): Number of steps in the denoising process. Default is 20.
  • numberOfSamples (string): Specifies how many samples to generate. Default is "1".
  • imageResolution (string): Resolution of the generated image (options: 256, 512, 768). Default is "512".
  • lowThreshold (integer): Lower threshold for Canny edge detection. Default is 100.
  • highThreshold (integer): Upper threshold for Canny edge detection. Default is 200.
  • additionalPrompt (string): Additional text appended to the main prompt to enhance quality. Defaults to 'best quality, extremely detailed.'
  • negativePrompt (string): Specifies undesirable characteristics or features for exclusion in the output. Defaults to common issues.

Example Input:

{
  "image": "https://replicate.delivery/pbxt/IKJO0Z6768YQahgAfgUF00iJCi2wPNVB8EwefQWodZagisYt/pose2.png",
  "scale": 9,
  "prompt": "an astronaut on the moon, digital art",
  "denoisingSteps": 20,
  "negativePrompt": "longbody, lowres, bad anatomy, bad hands, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality",
  "imageResolution": "512",
  "numberOfSamples": "1",
  "additionalPrompt": "best quality, extremely detailed",
  "detectionResolution": 512
}

Output

Upon successful execution, the action returns an array of URLs pointing to the generated images based on the input parameters.

Example Output:

[
  "https://assets.cognitiveactions.com/invocations/523b708e-4983-4a13-9a7a-5cc8d05fbaa5/a79d9681-46eb-45ed-8d4e-13c8c3c88177.png",
  "https://assets.cognitiveactions.com/invocations/523b708e-4983-4a13-9a7a-5cc8d05fbaa5/5264a6c8-cf0f-452b-bc02-e7cc2415eaab.png"
]

Conceptual Usage Example (Python)

Here is a conceptual Python code snippet that demonstrates how to call this Cognitive 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 = "8882ce4a-9c2b-4eb5-95c7-fdb47a6f495f"  # Action ID for Modify Images Using Pose Detection

# Construct the input payload based on the action's requirements
payload = {
    "image": "https://replicate.delivery/pbxt/IKJO0Z6768YQahgAfgUF00iJCi2wPNVB8EwefQWodZagisYt/pose2.png",
    "scale": 9,
    "prompt": "an astronaut on the moon, digital art",
    "denoisingSteps": 20,
    "negativePrompt": "longbody, lowres, bad anatomy, bad hands, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality",
    "imageResolution": "512",
    "numberOfSamples": "1",
    "additionalPrompt": "best quality, extremely detailed",
    "detectionResolution": 512
}

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 example, replace YOUR_COGNITIVE_ACTIONS_API_KEY with your actual API key. The action_id corresponds to the Modify Images Using Pose Detection action. The input payload is structured according to the specifications, allowing for a smooth interaction with the API.

Conclusion

The jagilley/controlnet-pose Cognitive Action for modifying images using pose detection opens up exciting possibilities for developers looking to enhance their applications. By leveraging the power of pose detection and image generation, you can create unique and compelling visual content. Start exploring these capabilities today and consider how they can elevate your projects to the next level!

← Back to All Articles