3d-reconstruction

Transform Images into 3D Geometry with Geowizard

25 Apr 2025
Transform Images into 3D Geometry with Geowizard

Geowizard is an innovative tool that empowers developers to harness the potential of advanced image processing through its Cognitive Actions. One of its standout features is the ability to estimate 3D geometry from a single image. This capability is particularly beneficial as it simplifies the process of creating three-dimensional representations, making it accessible for various applications such as gaming, virtual reality, architecture, and product visualization. By utilizing Diffusion Priors, Geowizard enhances prediction accuracy in diverse environments, including indoor, outdoor, and object settings.

Imagine being able to take a flat image and extract its three-dimensional structure, all with a few simple API calls. This action opens up numerous possibilities for developers looking to integrate 3D modeling into their applications without the need for complex setups or specialized knowledge in geometry processing.

Prerequisites

To get started with Geowizard's Cognitive Actions, you will need an API key to authenticate your requests. Familiarity with general API calls will also be beneficial for seamless integration.

Estimate 3D Geometry from Single Image

The "Estimate 3D Geometry from Single Image" action is designed to transform a two-dimensional image into a three-dimensional representation, addressing the challenge of reconstructing 3D models from flat images. This action is part of the 3D reconstruction category and is pivotal for applications that require depth perception and spatial understanding.

Input Requirements

The input for this action is a JSON object that must include:

  • inputImage: A URI pointing to the image that needs processing (required).
  • seed: An optional integer that determines the random seed for reproducibility (default is 123).
  • domain: Specifies the content category (indoor, outdoor, or object), with a default of "indoor".
  • ensembleSize: An integer ranging from 1 to 15 that indicates the number of models to ensemble (default is 4).
  • denoisingSteps: An integer specifying the number of steps for the denoising process, ranging from 1 to 50 (default is 10).
  • processingResolution: An integer that sets the resolution for processing, either 0 or 768 (default is 768).

Example Input

{
  "seed": 123,
  "domain": "indoor",
  "inputImage": "https://replicate.delivery/pbxt/KdKJagGL6FVq0mz0pFsQsR1egZWvWlxcvVVh8KlyaeEbt0Es/basketball_dog.jpg",
  "ensembleSize": 4,
  "denoisingSteps": 10,
  "processingResolution": 768
}

Expected Output

Upon processing, the action will return a list of URIs pointing to the generated 3D geometry images. This output allows developers to visualize the estimated 3D structure derived from the input image.

Example Output

[
  "https://assets.cognitiveactions.com/invocations/56701b1f-b529-43a4-83a5-ae425561ba64/2f757723-91ba-451a-b0e5-82aaa03d9786.png",
  "https://assets.cognitiveactions.com/invocations/56701b1f-b529-43a4-83a5-ae425561ba64/16049924-a6f3-4c59-90ff-a3e82e513d56.png"
]

Use Cases for this Action

  • Gaming Development: Enhance gaming environments by generating 3D models from concept art or real-world images, allowing for richer immersive experiences.
  • Virtual Reality: Create realistic 3D environments for VR applications by converting user-uploaded images into 3D spaces.
  • E-Commerce: Provide customers with a 3D view of products by transforming product images into 3D models, improving the online shopping experience.
  • Architecture and Design: Assist architects and designers in visualizing spaces by creating 3D representations from blueprints or sketches.
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 = "c497fe70-25be-4bd8-94fc-e0592413a601" # Action ID for: Estimate 3D Geometry from Single Image

# Construct the exact input payload based on the action's requirements
# This example uses the predefined example_input for this action:
payload = {
  "seed": 123,
  "domain": "indoor",
  "inputImage": "https://replicate.delivery/pbxt/KdKJagGL6FVq0mz0pFsQsR1egZWvWlxcvVVh8KlyaeEbt0Es/basketball_dog.jpg",
  "ensembleSize": 4,
  "denoisingSteps": 10,
  "processingResolution": 768
}

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

Geowizard's ability to estimate 3D geometry from a single image represents a significant advancement in image processing technology. By integrating this action into your applications, you can unlock new dimensions of interactivity and visualization, enhancing user engagement and experience. Whether you are in gaming, e-commerce, or design, the possibilities are vast. Start exploring Geowizard today to bring your images to life in three dimensions!

← Back to All Articles