Text-to-3D Generation with Bidirectional Diffusion Using Both 2D and 3D Priors

≈ 20min A yellow and green oil painting style eagle head (c) ProlificDreamer (b) Zero-123 (a) Shap-E Figure 1. Our BiDiff can efficiently generate high-quality 3D objects. It alleviates all these issues in previous 3D generative models: (a) low-texture quality, (b) multi-view inconsistency, and (c) geometric incorrectness (e.g., multi-face Janus problem). The outputs of our model can be further combined with optimization-based methods (e.g., ProlificDreamer) to generate better 3D geometries with slightly longer processing time (bottom row).