TY  - JOUR
TI  - DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generation
AU  - Ruiz, N.
AU  - Li, Y.
AU  - Jampani, V.
AU  - Pritch, Y.
AU  - Rubinstein, M.
AU  - Aberman, K.
PY  - 2023
DA  - 2023///
JO  - CVPR 2023 (IEEE/CVF)
AB  - Ruiz et al. (2022, published CVPR 2023) introduce DreamBooth, a fine-tuning method that binds a unique text identifier to a specific subject by fine-tuning all weights of a diffusion model on a small reference image set, with a prior-preservation loss that prevents language drift. The result allows the subject to be synthesised in arbitrary contexts, poses, and scenes while maintaining identity. DreamBooth was the first practically accessible character-lock-in method and has been widely productised in animation pre-production pipelines for consistent character generation.
KW  - generative-ai
KW  - image-generation
KW  - character-animation
KW  - training-data
UR  - https://arxiv.org/abs/2208.12242
LA  - en
ER  -