Create 3D reconstructions from images using MASt3R and 3DGS
Generate 3D content from images or text
Gradio demo of CharacterGen (SIGGRAPH 2024)
Generate 3D room layouts from RGB panoramas
Explore Minnesota with a 3D video map
Create 3D models from images using depth estimation
Play an interactive 3D Pong game
Scalable and Versatile 3D Generation from images
Create an interactive GTA-style city with collisions and NPC dialogue
3D Generation from text prompts
Generate 3D models from images
Explore a 3D map of Minnesota
The AniMer Demo
MASt3r+3DGS is a powerful 3D modeling tool designed to create detailed 3D reconstructions from 2D images. By combining the capabilities of MASt3r and 3DGS, this tool enables users to generate accurate and high-resolution 3D models from multi-view image data. It is particularly useful for applications in computer vision, architectural modeling, and game development.
• Multi-View Image Processing: Handles multiple images from different angles to reconstruct 3D models.
• Advanced 3D Reconstruction: Combines MASt3r (Multi-View Stereo) and 3DGS (3D Geometry Solver) for precise modeling.
• High-Resolution Output: Produces detailed 3D models with realistic textures and surfaces.
• Iterative Refinement: refines models through multiple passes for optimal accuracy.
• Support for Various Formats: Exports models in standard formats (e.g., OBJ, STL, PLY) for compatibility with other software.
What types of images can MASt3r+3DGS process?
MASt3r+3DGS works best with high-resolution, well-lit images taken from multiple angles. It supports common formats like JPG, PNG, and BMP.
How long does the reconstruction process take?
Reconstruction time depends on the number of images and the complexity of the scene. Simple scenes may take minutes, while detailed scenes can take hours or more.
Can I use MASt3r+3DGS for real-time 3D modeling?
No, MASt3r+3DGS is designed for offline processing. It is optimized for generating high-quality 3D models rather than real-time applications.