Research map / Image generation / Text-to-image diffusion
DreamBooth3D: Subject-Driven Text-to-3D Generation
Amit Raj, Srinivas Kaza, Ben Poole, Michael Niemeyer, Nataniel Ruiz, Ben Mildenhall, Shiran Zada, Kfir Aberman, Michael Rubinstein, Jonathan Barron, Yuanzhen Li, Varun Jampani
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 |
|---|---|---|---|
| use 3-stage optimization (partial DreamBooth NeRF, pseudo multi-view data, multi-view DreamBooth NeRF) instead of naive DreamBooth followed by DreamFusion | DreamBooth3D · DreamBooth dataset, CLIP R-Precision | CLIP R-Precision ViT-B/16: 0.509 -> 0.783 | Improved |
| use 3-stage optimization instead of naive DreamBooth+Fusion | DreamBooth3D · DreamBooth dataset, CLIP R-Precision | CLIP R-Precision ViT-L-14: 0.506 -> 0.797 | Improved |
| fully finetune DreamBooth before DreamFusion (naive DreamBooth+Fusion) (ablation) | DreamBooth+Fusion · DreamBooth dataset, qualitative | 3D consistency (Janus problem) | Hurt |
| use partially finetuned DreamBooth (150 iterations) for the initial NeRF (ablation) | DreamBooth3D stage 1 · DreamBooth dataset, qualitative | 3D coherence | Improved |
| run stage 3 (multi-view DreamBooth plus final NeRF) after the initial NeRF (ablation) | DreamBooth3D · DreamBooth dataset, qualitative | subject likeness | Improved |
| add weak reconstruction loss on pseudo multi-view images in final NeRF (ablation) | DreamBooth3D stage 3 · DreamBooth dataset, qualitative | color fidelity | Improved |
| DreamBooth3D versus DreamBooth+Fusion and Latent-NeRF in pairwise user study | DreamBooth3D · DreamBooth dataset user study | preference on 3D consistency, subject fidelity, prompt fidelity | Improved |
Closest papers by meaning
- DIRECT-3D: Learning Direct Text-to-3D Generation on Massive Noisy 3D Data (CVPR 2024 · similarity 0.87)
- AttnDreamBooth: Towards Text-Aligned Personalized Text-to-Image Generation (NeurIPS 2024 · similarity 0.87)
- DreamDPO: Aligning Text-to-3D Generation with Human Preferences via Direct Preference Optimization (ICML 2025 · similarity 0.87)
- DreamAlign: Dynamic Text-to-3D Optimization with Human Preference Alignment (AAAI 2025 · similarity 0.87)
- A Unified Approach for Text- and Image-guided 4D Scene Generation (CVPR 2024 · similarity 0.87)
- ViewDiff: 3D-Consistent Image Generation with Text-to-Image Models (CVPR 2024 · similarity 0.87)
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.
