3d-reconstruction

Create Stunning 3D Textured Assets with Hunyuan3d 2

27 Apr 2025
Create Stunning 3D Textured Assets with Hunyuan3d 2

In the world of 3D modeling and asset creation, having the right tools can significantly enhance the quality and efficiency of your projects. Hunyuan3d 2 offers a suite of Cognitive Actions that empower developers to generate high-resolution textured 3D assets from simple input images. By leveraging advanced diffusion models, this service simplifies the asset creation process while ensuring detailed and reproducible outputs.

Imagine being able to transform a 2D image into a fully textured 3D model with just a few parameters. This capability is invaluable across various industries, including gaming, animation, and virtual reality, where high-quality assets are crucial. Whether you're developing a new game environment, creating assets for a film, or designing interactive experiences, Hunyuan3d 2 can help streamline your workflow and elevate your creative projects.

Prerequisites

To get started with Hunyuan3d 2, you'll need a Cognitive Actions API key and a basic understanding of API calls.

Generate High Resolution 3D Textured Assets

The "Generate High Resolution 3D Textured Assets" action is designed to create detailed 3D models from 2D images, making it easier for developers to produce high-quality assets for various applications. This action addresses the challenge of converting flat images into three-dimensional representations, allowing for more dynamic and engaging content.

Input Requirements

To utilize this action, the following input parameters are required:

  • Seed (integer): A random seed used for generation, ensuring reproducibility. Default is 1234.
  • Image (URI): The URI of the input image that will be used to generate the 3D shape. The image must be accessible over the internet.
  • Steps (integer): The number of inference steps taken during generation. The valid range is from 20 to 50, with a default of 50.
  • Guidance Scale (number): This parameter influences how closely the output adheres to the input data. The range is from 1 to 20, with a default of 5.5.
  • Octree Resolution (integer): This determines the detail level of the output mesh, with options of 256, 384, or 512. Default is 512.
  • Remove Background (boolean): Indicates whether to remove the background from the input image, with a default value of true.

Expected Output

Upon successful execution, the action will return a link to the generated 3D mesh, which can be utilized in various applications, such as gaming or virtual reality experiences.

Example output:

{
  "mesh": "https://assets.cognitiveactions.com/invocations/e3059932-4dfc-49a0-a37e-1a0b716d94e4/e6629dd7-33b6-444e-8095-561260290182.glb"
}

Use Cases for this Specific Action

  • Game Development: Quickly generate high-quality 3D assets for characters, environments, or props, reducing the time spent on manual modeling.
  • Virtual Reality Experiences: Create immersive environments with detailed assets that enhance user engagement and realism.
  • Animation: Develop richly textured models for animated films or shorts, ensuring that characters and scenes look polished and professional.
  • Prototyping: Use generated assets in product design or architectural visualizations, allowing for rapid iteration and visualization of concepts.
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 = "0a2ce5d2-6395-4a54-8e69-390a8fd76403" # Action ID for: Generate High Resolution 3D Textured Assets

# Construct the exact input payload based on the action's requirements
# This example uses the predefined example_input for this action:
payload = {
  "seed": 1234,
  "image": "https://replicate.delivery/pbxt/MVC2B2XKgv4X13qIpW6t2m59EVfY2CqaS9e2CSsWNHPJjQAd/image.png",
  "steps": 50,
  "guidanceScale": 5.5,
  "octreeResolution": 512,
  "removeBackground": true
}

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

Hunyuan3d 2 provides a powerful solution for developers looking to create stunning 3D textured assets efficiently. By transforming 2D images into high-resolution 3D models, it opens up a world of possibilities for gaming, animation, and virtual reality projects. With customizable input parameters, you can fine-tune the generation process to meet your specific needs.

To leverage the full potential of Hunyuan3d 2, start integrating these actions into your workflow today and elevate your creative projects to new heights!

← Back to All Articles