Research map / Image generation / Text-to-image diffusion
Diverse Text-to-3D Synthesis with Augmented Text Embedding
Uy Dieu Tran, Minh N. Hoang Luu, Phong Ha Nguyen, Khoi Nguyen, Binh-Son Hua
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| condition each particle on augmented prompts via HiPer textual inversion of reference images (ablation) | VSD / text-to-3D · Text-to-3D prompts | Cosine Sim: 0.476 -> 0.355 | Improved |
| add HiPer augmented prompts to VSD (ablation) | VSD · Text-to-3D prompts | Inception Variance (IV): 4.586 -> 5.109 | Improved |
| add HiPer augmented prompts to VSD (ablation) | VSD · Text-to-3D prompts | Inception Quality (IQ): 3.345 -> 3.662 | Hurt |
| replace LoRA domain adapter with shared learnable tokens (ablation) | TSD · Text-to-3D prompts | Training time (hours): 9.50 -> 7.16 | Improved |
| replace LoRA with shared learnable tokens (ablation) | TSD · Text-to-3D prompts | IQ / Cosine Sim: 3.662 -> 3.614 IQ; 0.355 -> 0.380 Cosine Sim | Hurt |
| increase number of HiPer tokens (ablation) | TSD · Frog wearing a sweater prompt | IQ: 3.375 -> 5.193 (1 to 20 tokens) | Hurt |
| increase number of HiPer tokens (ablation) | TSD · Frog wearing a sweater prompt | Cosine Sim / IV: 0.403 -> 0.425 Cosine Sim (1 to 20 tokens) | No clear effect |
| use reference-image inversion instead of ChatGPT-generated augmented prompts (ablation) | VSD particle conditioning · Qualitative (ice cream sundae) | fidelity and diversity | Improved |
| apply method with 3D Gaussian Splatting instead of Instant-NGP | TSD · 6 particles | training time: 7 hours -> 2 hours | Improved |
| TSD vs SDS | TSD · Text-to-3D prompts | Cosine Sim: 0.720 -> 0.380 | Improved |
Closest papers by meaning
- IT3D: Improved Text-to-3D Generation with Explicit View Synthesis (AAAI 2024 · similarity 0.91)
- Instant3D: Fast Text-to-3D with Sparse-view Generation and Large Reconstruction Model (ICLR 2024 · similarity 0.89)
- VP3D: Unleashing 2D Visual Prompt for Text-to-3D Generation (CVPR 2024 · similarity 0.89)
- EfficientDreamer: High-Fidelity and Robust 3D Creation via Orthogonal-view Diffusion Priors (CVPR 2024 · similarity 0.89)
- Let 2D Diffusion Model Know 3D-Consistency for Robust Text-to-3D Generation (ICLR 2024 · similarity 0.89)
- Exploring the Deep Fusion of Large Language Models and Diffusion Transformers for Text-to-Image Synthesis (CVPR 2025 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
