We've released the Code & Gradio demo & Colab demo for our paper, DATID-3D: Diversity-Preserved Domain Adaptation Using Text-to-Image Diffusion for 3D Generative Model (accepted to CVPR2023). We showcase the demo of text-guided manipulated 3D reconstuction beyond text-guided image manipulation!
- Author: Gwanghyun Kim and Se Young Chun @ SNU
- Paper: [ Ссылка ]
- Project: [ Ссылка ]
- Code & Colab Demo: [ Ссылка ]
DATID-3D succeeded in text-guided domain adaptation of 3D-aware generative models while preserving diversity that is inherent in the text prompt as well as enabling high-quality pose-controlled image synthesis with excellent text-image correspondence.
Ещё видео!