Text-driven 3D Human Generation via Contrastive Preference Optimization
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866909550928461824 |
|---|---|
| author | Zhou, Pengfei Shen, Xukun Hu, Yong |
| author_facet | Zhou, Pengfei Shen, Xukun Hu, Yong |
| contents | Recent advances in Score Distillation Sampling (SDS) have improved 3D human generation from textual descriptions. However, existing methods still face challenges in accurately aligning 3D models with long and complex textual inputs. To address this challenge, we propose a novel framework that introduces contrastive preferences, where human-level preference models, guided by both positive and negative prompts, assist SDS for improved alignment. Specifically, we design a preference optimization module that integrates multiple models to comprehensively capture the full range of textual features. Furthermore, we introduce a negation preference module to mitigate over-optimization of irrelevant details by leveraging static-dynamic negation prompts, effectively preventing ``reward hacking". Extensive experiments demonstrate that our method achieves state-of-the-art results, significantly enhancing texture realism and visual alignment with textual descriptions, particularly for long and complex inputs. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_08977 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Text-driven 3D Human Generation via Contrastive Preference Optimization Zhou, Pengfei Shen, Xukun Hu, Yong Computer Vision and Pattern Recognition Recent advances in Score Distillation Sampling (SDS) have improved 3D human generation from textual descriptions. However, existing methods still face challenges in accurately aligning 3D models with long and complex textual inputs. To address this challenge, we propose a novel framework that introduces contrastive preferences, where human-level preference models, guided by both positive and negative prompts, assist SDS for improved alignment. Specifically, we design a preference optimization module that integrates multiple models to comprehensively capture the full range of textual features. Furthermore, we introduce a negation preference module to mitigate over-optimization of irrelevant details by leveraging static-dynamic negation prompts, effectively preventing ``reward hacking". Extensive experiments demonstrate that our method achieves state-of-the-art results, significantly enhancing texture realism and visual alignment with textual descriptions, particularly for long and complex inputs. |
| title | Text-driven 3D Human Generation via Contrastive Preference Optimization |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2502.08977 |