We use the threestudio implementation for all the baselines. A fixed configuration is used for each method (including ours) without prompt-wise hyperparameter tuning. The results here are without cherry picking.
Dreamfusion-IF
Magic3D-IF-SD
Text2Mesh-IF
ProlificDreamer
Ours
Daenerys Targaryen from game of throne, full body, blender 3d, artstation and behance, Disney Pixar, Mobile game character, clash royale, cute
Dragon armor, 3D asset
an astronaut riding a horse
Viking axe, fantasy, weapon, blender, 8k, HD
Samurai koala bear
A bulldog wearing a black pirate hat
beautiful, intricate butterfly
a DSLR photo of an eggshell broken in two with an adorable chick standing next to it
Corgi riding a rocket
A crab, low poly
@article{shi2023MVDream,
author = {Shi, Yichun and Wang, Peng and Ye, Jianglong and Mai, Long and Li, Kejie and Yang, Xiao},
title = {MVDream: Multi-view Diffusion for 3D Generation},
journal = {arXiv:2308.16512},
year = {2023},
}