Transforming 2D Images into 3D Objects with Prolific Splats Cognitive Actions

24 Apr 2025
Transforming 2D Images into 3D Objects with Prolific Splats Cognitive Actions

In the world of visual computing, the ability to transform 2D images into 3D objects opens up a plethora of possibilities for developers. The ltejedor/prolific-splats API provides a powerful Cognitive Action called Convert Image to 3D that leverages advanced techniques such as DreamGaussian and Prolific Dreamer. This action not only enhances the depth and detail of images but also streamlines the process of creating 3D models from flat images, making it a valuable tool for applications in gaming, virtual reality, and design.

Prerequisites

Before you start using the Cognitive Actions, ensure you have the following:

  • An API key for the Cognitive Actions platform. This will allow you to authenticate your requests.
  • Basic knowledge of JSON and how to structure payloads for API calls.

Authentication typically works by passing your API key in the request headers, ensuring secure access to the Cognitive Actions.

Cognitive Actions Overview

Convert Image to 3D

The Convert Image to 3D action enables you to create a three-dimensional representation of a 2D image. It preprocesses the input image, sets its orientation, and iteratively refines the output for greater accuracy and detail.

  • Category: 3D Reconstruction

Input

To execute this action, you need to provide the following input fields:

{
  "image": "https://replicate.delivery/pbxt/MgLUhC1pWZoTVASuysNTE6hg1AW3N89ayLUVh90TO00HVs3R/out-0-chair.png",
  "elevation": 0,
  "imageSize": 256,
  "numberOfSteps": 500,
  "numberOfPointSamples": 1000,
  "numberOfRefinementSteps": 50
}
  • image: (string) URI of the input image used for 3D conversion. (Required)
  • elevation: (integer) Elevation angle of the input image, ranging from -90 to 90 degrees. (Optional, default is 0)
  • imageSize: (integer) Size to which the input image will be preprocessed. (Optional, default is 256 pixels)
  • numberOfSteps: (integer) Total number of iterations for the process. (Optional, default is 500)
  • numberOfPointSamples: (integer) Number of points sampled for Gaussian Splatting, between 200 and 10000. (Optional, default is 1000)
  • numberOfRefinementSteps: (integer) Number of refinement iterations applied during the process. (Optional, default is 50)

Output

Upon successful execution, the action returns a list of URLs pointing to the generated 3D object and associated files:

[
  "https://assets.cognitiveactions.com/invocations/74065e72-040c-4a39-8e56-98841fdc9010/b4fc44d4-37b8-4407-8c6b-c2b3d3f55454.zip",
  "https://assets.cognitiveactions.com/invocations/74065e72-040c-4a39-8e56-98841fdc9010/43b79bb2-e839-487a-b9c0-ebfa602e8061.png",
  "https://assets.cognitiveactions.com/invocations/74065e72-040c-4a39-8e56-98841fdc9010/d420ccae-ea00-427e-a279-c7c456cc489f.mp4"
]
  • The first URL typically points to a zip file containing the 3D model.
  • The second URL usually leads to a rendered image of the 3D object.
  • The third URL often provides a video showcasing the 3D model.

Conceptual Usage Example (Python)

Here's a conceptual Python code snippet to demonstrate how you might utilize this 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 for Convert Image to 3D
action_id = "f3257a3b-50df-447e-8426-f05383d13513"

# Construct the input payload based on the action's requirements
payload = {
    "image": "https://replicate.delivery/pbxt/MgLUhC1pWZoTVASuysNTE6hg1AW3N89ayLUVh90TO00HVs3R/out-0-chair.png",
    "elevation": 0,
    "imageSize": 256,
    "numberOfSteps": 500,
    "numberOfPointSamples": 1000,
    "numberOfRefinementSteps": 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 placeholder for the API key and adjust the endpoint as necessary.
  • The action ID is set for Convert Image to 3D.
  • The input payload is structured based on the action's requirements.

Conclusion

The Convert Image to 3D action from the ltejedor/prolific-splats API offers an impressive solution for developers looking to enrich their applications with 3D content. By transforming 2D images into detailed 3D models, you can enhance user experiences in gaming, virtual reality, and design applications. Consider exploring additional use cases where this action could be integrated to elevate your projects further.

← Back to All Articles