Scalable and Versatile 3D Generation from images
Create a dynamic 3D scene with random shapes and lights
Generate 3D models from text descriptions
Create 3D models from images using depth estimation
Image to 3D with DPT + 3D Point Cloud
Generate a dynamic 3D scene with floating lights and torus knots
text-to-3D & image-to-3D
Generate 3D scenes with dynamic lighting and shapes
Create 3D reconstructions from images using MASt3R and 3DGS
Generate 3D molecular models from SMILES strings
Create an interactive 3D sphere fountain that follows your mouse
Generate interactive 3D torus knots in a virtual environment
Generate a dynamic 3D scene with rotating lights and knots
TRELLIS is a scalable and versatile 3D modeling tool designed to generate high-quality 3D models directly from images. It leverages advanced AI technology to convert 2D images into 3D models, allowing users to extract GLB or Gaussian files for further use in various applications. This tool is ideal for artists, designers, and developers looking to bridge the gap between 2D and 3D content creation.
• Image-to-3D Conversion: Easily transform 2D images into detailed 3D models using AI-driven algorithms.
• Scalable Output: Generate 3D models at varying resolutions to suit different project requirements.
• Multiple Export Formats: Supports GLB and Gaussian file exports for compatibility with popular 3D software.
• Versatile Applications: Suitable for gaming, architecture, product design, and virtual reality projects.
• User-Friendly API: Integrate TRELLIS into your workflow with a simple and intuitive API.
What file formats does TRELLIS support for input?
TRELLIS supports standard image formats such as JPEG, PNG, and BMP for input.
Can I customize the output resolution of the 3D model?
Yes, TRELLIS allows you to adjust the resolution and complexity of the generated 3D model to meet your specific requirements.
Is TRELLIS suitable for real-time 3D generation?
TRELLIS is optimized for scalability but may require additional optimizations for real-time applications depending on the use case.