notesum.ai

Published at December 3

Viewpoint Consistency in 3D Generation via Attention and CLIP Guidance

cs.CV

Released Date: December 3, 2024

Authors: Qing Zhang¹, Zehao Chen, Jinguang Tong, Jing Zhang, Jie Hong, Xuesong Li

Aff.: ¹Australian National University

Arxiv: http://arxiv.org/pdf/2412.02287v1

Refer to caption

Method	JR(%) $\downarrow$	View-CS(%) $\uparrow$
LucidDreamer [18]	80.00	29.40
LucidDreamer+ACG (Ours)	30.00	30.26
Magic3D [19]	75.00	28.20
Magic3D+ACG (Ours)	35.00	29.98
DreamFusion [27]	80.00	29.31
DreamFusion+ACG (Ours)	35.00	30.13