{"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/fastcomposer-tuning-free-multi-subject-image","title":"FastComposer: Tuning-Free Multi-Subject Image Generation with Localized Attention","arxiv_id":"2305.10431","date":"2023-05-17","proceeding":null,"authors":["Guangxuan Xiao","Tianwei Yin","William T. Freeman","Frédo Durand","Song Han"],"abstract":"Diffusion models excel at text-to-image generation, especially in subject-driven generation for personalized images. However, existing methods are inefficient due to the subject-specific fine-tuning, which is computationally intensive and hampers efficient deployment. Moreover, existing methods struggle with multi-subject generation as they often blend features among subjects. We present FastComposer which enables efficient, personalized, multi-subject text-to-image generation without fine-tuning. FastComposer uses subject embeddings extracted by an image encoder to augment the generic text conditioning in diffusion models, enabling personalized image generation based on subject images and textual instructions with only forward passes. To address the identity blending problem in the multi-subject generation, FastComposer proposes cross-attention localization supervision during training, enforcing the attention of reference subjects localized to the correct regions in the target images. Naively conditioning on subject embeddings results in subject overfitting. FastComposer proposes delayed subject conditioning in the denoising step to maintain both identity and editability in subject-driven image generation. FastComposer generates images of multiple unseen individuals with different styles, actions, and contexts. It achieves 300$\\times$-2500$\\times$ speedup compared to fine-tuning-based methods and requires zero extra storage for new subjects. FastComposer paves the way for efficient, personalized, and high-quality multi-subject image creation. Code, model, and dataset are available at https://github.com/mit-han-lab/fastcomposer.","url_abs":"https://arxiv.org/abs/2305.10431v2","url_pdf":"https://arxiv.org/pdf/2305.10431v2.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":"fastcomposer-tuning-free-multi-subject-image","repo_url":"https://github.com/mit-han-lab/fastcomposer","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"denoising","task_name":"Denoising"},{"task_slug":"diffusion-personalization","task_name":"Diffusion Personalization"},{"task_slug":"diffusion-personalization-tuning-free","task_name":"Diffusion Personalization Tuning Free"},{"task_slug":"image-generation","task_name":"Image Generation"},{"task_slug":"personalized-image-generation","task_name":"Personalized Image Generation"},{"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/diffusion-personalization-tuning-free-on","task":"Diffusion Personalization Tuning Free","dataset":"AgeDB","model":"FastComposer","rank_in_archive_order":7,"of":7,"metrics":{"Cosine Similarity":"0.256","FID":"31.736","LPIPS":"0.487"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2305.10431","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2305.10431"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/mit-han-lab/fastcomposer","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran":6,"unverified":11},"by_repo_kind":{"official":{"samples":17,"ran":6,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"addaaf66c3ab7352","entry":"get_test_transforms","repo":"mit-han-lab/fastcomposer","repo_kind":"official","path":"fastcomposer/transforms.py","file_url":"https://github.com/mit-han-lab/fastcomposer/blob/HEAD/fastcomposer/transforms.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"addaaf66c3ab7352"}},{"code_sha256_prefix":"c036796b62b5952d","entry":"get_train_transforms","repo":"mit-han-lab/fastcomposer","repo_kind":"official","path":"fastcomposer/transforms.py","file_url":"https://github.com/mit-han-lab/fastcomposer/blob/HEAD/fastcomposer/transforms.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"c036796b62b5952d"}},{"code_sha256_prefix":"15ed5aec7f6bd270","entry":"get_train_transforms_with_segmap","repo":"mit-han-lab/fastcomposer","repo_kind":"official","path":"fastcomposer/transforms.py","file_url":"https://github.com/mit-han-lab/fastcomposer/blob/HEAD/fastcomposer/transforms.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"15ed5aec7f6bd270"}},{"code_sha256_prefix":"817e9d69c2cd23d3","entry":"prepare_image_token_idx","repo":"mit-han-lab/fastcomposer","repo_kind":"official","path":"fastcomposer/data.py","file_url":"https://github.com/mit-han-lab/fastcomposer/blob/HEAD/fastcomposer/data.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"817e9d69c2cd23d3"}},{"code_sha256_prefix":"ff242689fdee9a2e","entry":"scatter_object_embeddings","repo":"mit-han-lab/fastcomposer","repo_kind":"official","path":"fastcomposer/model.py","file_url":"https://github.com/mit-han-lab/fastcomposer/blob/HEAD/fastcomposer/model.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"ff242689fdee9a2e"}},{"code_sha256_prefix":"d3c9a29ac1be57e9","entry":"stable_diffusion_call_with_references_delayed_conditioning","repo":"mit-han-lab/fastcomposer","repo_kind":"official","path":"fastcomposer/pipeline.py","file_url":"https://github.com/mit-han-lab/fastcomposer/blob/HEAD/fastcomposer/pipeline.py","link_basis":"plan_row","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"d3c9a29ac1be57e9"}},{"code_sha256_prefix":"1b1c247142456689","entry":"collate_fn","repo":"mit-han-lab/fastcomposer","repo_kind":"official","path":"fastcomposer/data.py","file_url":"https://github.com/mit-han-lab/fastcomposer/blob/HEAD/fastcomposer/data.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"1b1c247142456689"}},{"code_sha256_prefix":"b77a02409cff8d05","entry":"compute_similarity_matrix","repo":"mit-han-lab/fastcomposer","repo_kind":"official","path":"evaluation/single_object/single_object_evaluation.py","file_url":"https://github.com/mit-han-lab/fastcomposer/blob/HEAD/evaluation/single_object/single_object_evaluation.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"b77a02409cff8d05"}},{"code_sha256_prefix":"2fa8e59bc1db947b","entry":"fuse_object_embeddings","repo":"mit-han-lab/fastcomposer","repo_kind":"official","path":"fastcomposer/model.py","file_url":"https://github.com/mit-han-lab/fastcomposer/blob/HEAD/fastcomposer/model.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"2fa8e59bc1db947b"}},{"code_sha256_prefix":"032366a6a60db29a","entry":"get_accessory_prompts","repo":"mit-han-lab/fastcomposer","repo_kind":"official","path":"evaluation/single_object/data.py","file_url":"https://github.com/mit-han-lab/fastcomposer/blob/HEAD/evaluation/single_object/data.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"032366a6a60db29a"}},{"code_sha256_prefix":"4876d90a82330a60","entry":"get_context_prompts","repo":"mit-han-lab/fastcomposer","repo_kind":"official","path":"evaluation/single_object/data.py","file_url":"https://github.com/mit-han-lab/fastcomposer/blob/HEAD/evaluation/single_object/data.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"4876d90a82330a60"}},{"code_sha256_prefix":"c39580137a368949","entry":"get_data_loader","repo":"mit-han-lab/fastcomposer","repo_kind":"official","path":"fastcomposer/data.py","file_url":"https://github.com/mit-han-lab/fastcomposer/blob/HEAD/fastcomposer/data.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"c39580137a368949"}},{"code_sha256_prefix":"8d486087a7446edf","entry":"get_style_prompts","repo":"mit-han-lab/fastcomposer","repo_kind":"official","path":"evaluation/single_object/data.py","file_url":"https://github.com/mit-han-lab/fastcomposer/blob/HEAD/evaluation/single_object/data.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"8d486087a7446edf"}},{"code_sha256_prefix":"20c814d007b3b03b","entry":"greedy_matching","repo":"mit-han-lab/fastcomposer","repo_kind":"official","path":"evaluation/single_object/single_object_evaluation.py","file_url":"https://github.com/mit-han-lab/fastcomposer/blob/HEAD/evaluation/single_object/single_object_evaluation.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"20c814d007b3b03b"}},{"code_sha256_prefix":"15ad961bd4dd81e7","entry":"parse_args","repo":"mit-han-lab/fastcomposer","repo_kind":"official","path":"fastcomposer/utils.py","file_url":"https://github.com/mit-han-lab/fastcomposer/blob/HEAD/fastcomposer/utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"15ad961bd4dd81e7"}},{"code_sha256_prefix":"09e9aa9e75519b16","entry":"read_reference_images","repo":"mit-han-lab/fastcomposer","repo_kind":"official","path":"evaluation/single_object/single_object_evaluation.py","file_url":"https://github.com/mit-han-lab/fastcomposer/blob/HEAD/evaluation/single_object/single_object_evaluation.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"09e9aa9e75519b16"}},{"code_sha256_prefix":"0a5a54048f832e12","entry":"unet_store_cross_attention_scores","repo":"mit-han-lab/fastcomposer","repo_kind":"official","path":"fastcomposer/model.py","file_url":"https://github.com/mit-han-lab/fastcomposer/blob/HEAD/fastcomposer/model.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"0a5a54048f832e12"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}