Large Multi-View Gaussian Model

This tutorial is a demo implementation of LGM. LGM, or Large Multi-View Gaussian Model, is an innovative framework for generating high-resolution 3D models from textual cues or single-view images. It was first proposed by researchers from Peking University, Nanyang Technological University S-Lab, and Shanghai Artificial Intelligence Laboratory in the paper “LGM: Large Multi-View Gaussian Model for High-Resolution 3D Content CreationThe LGM framework uses multi-view Gaussian features as 3D representation and asymmetric U-Net as the backbone network to achieve high-fidelity and efficient 3D model generation. This method can generate 3D objects within 5 seconds and increase the training resolution to 512, thereby achieving high-resolution 3D content generation.

Effect display

Run steps

1. After cloning the tutorial container and successfully starting it, follow the instructions in the figure below to enter the operation page:

2. Upload a picture/enter a prompt word/a combination of the two to generate a 3D display effect:

Exchange and discussion

🖌️ If you see a high-quality project, please leave a message in the background to recommend it! In addition, we have also established a tutorial exchange group. Welcome friends to scan the QR code and remark [SD Tutorial] to join the group to discuss various technical issues and share application effects↓