search

iSEE-Laboratory / DreamView

(ECCV 2024) Official implementation of Paper ''DreamView: Injecting View-specific Text Guidance into Text-to-3D Generation''
MIT License
33 stars 0 forks source link

How can I generate a multi-view or 3D result by providing a prompt and a corresponding image? #1

Closed 2019211753 closed 2 months ago

2019211753 commented 2 months ago

Thank you for your amazing work and active contributions to open source. I'm very glad to receive your reply!

yanjk3 commented 2 months ago

Thanks for your interest. But, actually our DreamView is a text-to-multi-view (3D) generation method and thus can't generate multi-view images or 3D under an image prompt. If you want to generate multi-view images or 3D with an image prompt, you'd better refer to SyncDreamer, Zero-1-to-3, etc. And if you want to generate any multi-view images with your text prompts, you can modify Line 113 - Line 118 here, ways to generate 3D with text prompts will be updated later this month. Hope you have fun~~~~🥳🥳🥳

2019211753 commented 2 months ago

Thank you for your assistance. Your input was exactly what I needed.