{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/blip-diffusion-pre-trained-subject-1","title":"BLIP-Diffusion: Pre-trained Subject Representation for Controllable Text-to-Image Generation and Editing","arxiv_id":"2305.14720","date":"2023-05-24","proceeding":"NeurIPS 2023 11","authors":["Dongxu Li","Junnan Li","Steven C. H. Hoi"],"abstract":"Subject-driven text-to-image generation models create novel renditions of an input subject based on text prompts. Existing models suffer from lengthy fine-tuning and difficulties preserving the subject fidelity. To overcome these limitations, we introduce BLIP-Diffusion, a new subject-driven image generation model that supports multimodal control which consumes inputs of subject images and text prompts. Unlike other subject-driven generation models, BLIP-Diffusion introduces a new multimodal encoder which is pre-trained to provide subject representation. We first pre-train the multimodal encoder following BLIP-2 to produce visual representation aligned with the text. Then we design a subject representation learning task which enables a diffusion model to leverage such visual representation and generates new subject renditions. Compared with previous methods such as DreamBooth, our model enables zero-shot subject-driven generation, and efficient fine-tuning for customized subject with up to 20x speedup. We also demonstrate that BLIP-Diffusion can be flexibly combined with existing techniques such as ControlNet and prompt-to-prompt to enable novel subject-driven generation and editing applications. Code and models will be released at https://github.com/salesforce/LAVIS/tree/main/projects/blip-diffusion. Project page at https://dxli94.github.io/BLIP-Diffusion-website/.","url_abs":"https://arxiv.org/abs/2305.14720v2","url_pdf":"https://arxiv.org/pdf/2305.14720v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"blip-diffusion-pre-trained-subject-1","repo_url":"https://github.com/salesforce/lavis","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"BSD-3-Clause"}}],"tasks":[{"task_slug":"image-generation","task_name":"Image Generation"},{"task_slug":"personalized-image-generation","task_name":"Personalized Image Generation"},{"task_slug":"representation-learning","task_name":"Representation Learning"},{"task_slug":"text-to-image-generation-1","task_name":"Text to Image Generation"},{"task_slug":"text-to-image-generation","task_name":"Text-to-Image Generation"}],"methods":[{"method_slug":"diffusion","method_name":"Diffusion"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/personalized-image-generation-on-dreambench","task":"Personalized Image Generation","dataset":"DreamBooth","model":"BLIP-Diffusion SD v1.5","rank_in_archive_order":6,"of":7,"metrics":{"Concept Preservation (CP)":"0.547","Overall (CP * PF)":"0.271","Prompt Following (PF)":"0.495"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2305.14720","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}