Apply Hierarchical-Chain-of-Generation to Complex Attributes Text-to-3D Generation

Input Text: A man in black coat, yellow shirt, pink trousers, blue shoes and green hat is waving black Cross-attention map A man in black coat, yellow shirt, pink trousers, blue shoes and green hat is waving (a) Visualization of cross-attention map.

For longer text prompts, 2D Stable Diffusion (SD) [26] fails to accurately associate the word "black" with the correct spatial location in the generated image. This limitation poses a challenge for methods [15,27] lifting 2D to 3D using SD effectively.

Recent text-to-3D generation models have demonstrated remarkable abilities in producing high-quality 3D assets. Despite their great advancements, current models struggle to generate satisfying 3D objects with complex attributes. The difficulty for such complex attributes 3D generation arises from two aspects: (1) existing text-to-3D approaches typi- * Corresponding author.

cally lift text-to-image models to extract semantics via text encoders, while the text encoder exhibits limited comprehension ability for long descriptions, leading to deviated cross-attention focus, subsequently wrong attribute binding in generated results. (2) Objects with complex attributes often exhibit occlusion relationships between different parts, which demands a reasonable generation order as well as explicit disentanglement of different parts to enable structural coherent and attribute following results. Though some This CVPR paper is the Open Access version, provided by the Computer Vision Foundation.

Except for this watermark, it is identical to the accepted version; the final published version of the proceedings is available on IEEE Xplore.