{"about":{"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.","site":"https://codewithpapers.app","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","syntology":{"site":"https://syntology.ai","developers":"https://syntology.ai/developers","mcp":{"server":"https://syntology.ai/mcp","transport":"streamable-http","server_card":"https://syntology.ai/.well-known/mcp/server-card.json","auth":{"type":"trial token, no account","trial_token":"https://syntology.ai/api/oauth/trial/token","method":"POST","docs":"https://syntology.ai/developers"}},"have":"https://syntology.ai/api/graph/have?x=<method, arXiv id or title> (free, answers coverage only)","paper_base":"https://syntology.ai/paper/","atlas_base":"https://app.syntology.ai/?focus="},"machine_readable":[{"url":"https://codewithpapers.app/llms.txt","what":"the machine catalog: every machine-readable file, counted"},{"url":"https://codewithpapers.app/index/manifest.json","what":"paper-to-code index by arXiv id, with Syntology's counts"},{"url":"https://codewithpapers.app/search/manifest.json","what":"site search index (titles, authors) and its files"},{"url":"https://codewithpapers.app/download","what":"bulk files: Syntology's layer, described there"},{"url":"https://codewithpapers.app/build_manifest.json","what":"the build record: inputs, counts, exclusions, probes"}]},"url":"/paper/stimulating-the-diffusion-model-for-image","title":"Stimulating Diffusion Model for Image Denoising via Adaptive Embedding and Ensembling","arxiv_id":"2307.03992","date":"2023-07-08","proceeding":null,"authors":["Tong Li","Hansen Feng","Lizhi Wang","Zhiwei Xiong","Hua Huang"],"abstract":"Image denoising is a fundamental problem in computational photography, where achieving high perception with low distortion is highly demanding. Current methods either struggle with perceptual quality or suffer from significant distortion. Recently, the emerging diffusion model has achieved state-of-the-art performance in various tasks and demonstrates great potential for image denoising. However, stimulating diffusion models for image denoising is not straightforward and requires solving several critical problems. For one thing, the input inconsistency hinders the connection between diffusion models and image denoising. For another, the content inconsistency between the generated image and the desired denoised image introduces distortion. To tackle these problems, we present a novel strategy called the Diffusion Model for Image Denoising (DMID) by understanding and rethinking the diffusion model from a denoising perspective. Our DMID strategy includes an adaptive embedding method that embeds the noisy image into a pre-trained unconditional diffusion model and an adaptive ensembling method that reduces distortion in the denoised image. Our DMID strategy achieves state-of-the-art performance on both distortion-based and perception-based metrics, for both Gaussian and real-world image denoising.The code is available at https://github.com/Li-Tong-621/DMID.","url_abs":"https://arxiv.org/abs/2307.03992v5","url_pdf":"https://arxiv.org/pdf/2307.03992v5.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":"stimulating-the-diffusion-model-for-image","repo_url":"https://github.com/li-tong-621/dmid","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"color-image-denoising","task_name":"Color Image Denoising"},{"task_slug":"denoising","task_name":"Denoising"},{"task_slug":"image-denoising","task_name":"Image Denoising"}],"methods":[{"method_slug":"diffusion","method_name":"Diffusion"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/color-image-denoising-on-cbsd68-sigma100","task":"Color Image Denoising","dataset":"CBSD68 sigma100","model":"DMID-p","rank_in_archive_order":1,"of":2,"metrics":{"LPIPS":"0.208","PSNR":"23.94","SSIM":"0.7840"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-cbsd68-sigma100","task":"Color Image Denoising","dataset":"CBSD68 sigma100","model":"DMID-d","rank_in_archive_order":2,"of":2,"metrics":{"LPIPS":"0.283","PSNR":"25.96","SSIM":"0.8413"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-cbsd68-sigma15","task":"Color Image Denoising","dataset":"CBSD68 sigma15","model":"DMID-d","rank_in_archive_order":3,"of":10,"metrics":{"PSNR":"34.45"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-cbsd68-sigma150","task":"Color Image Denoising","dataset":"CBSD68 sigma150","model":"DMID-p","rank_in_archive_order":1,"of":2,"metrics":{"LPIPS":"0.264","PSNR":"22.47","SSIM":"0.7314"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-cbsd68-sigma150","task":"Color Image Denoising","dataset":"CBSD68 sigma150","model":"DMID-d","rank_in_archive_order":2,"of":2,"metrics":{"LPIPS":"0.361","PSNR":"24.54","SSIM":"0.8001"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-cbsd68-sigma200","task":"Color Image Denoising","dataset":"CBSD68 sigma200","model":"DMID-p","rank_in_archive_order":1,"of":2,"metrics":{"LPIPS":"0.312","PSNR":"21.37","SSIM":"0.6842"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-cbsd68-sigma200","task":"Color Image Denoising","dataset":"CBSD68 sigma200","model":"DMID-d","rank_in_archive_order":2,"of":2,"metrics":{"LPIPS":"0.392","PSNR":"23.57","SSIM":"0.7686"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-cbsd68-sigma25","task":"Color Image Denoising","dataset":"CBSD68 sigma25","model":"DMID-d","rank_in_archive_order":2,"of":9,"metrics":{"PSNR":"31.86"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-cbsd68-sigma250","task":"Color Image Denoising","dataset":"CBSD68 sigma250","model":"DMID-p","rank_in_archive_order":1,"of":2,"metrics":{"LPIPS":"0.352","PSNR":"20.77","SSIM":"0.6610"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-cbsd68-sigma250","task":"Color Image Denoising","dataset":"CBSD68 sigma250","model":"DMID-d","rank_in_archive_order":2,"of":2,"metrics":{"LPIPS":"0.455","PSNR":"22.88","SSIM":"0.7462"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-cbsd68-sigma50","task":"Color Image Denoising","dataset":"CBSD68 sigma50","model":"DMID-d (MMSE)","rank_in_archive_order":3,"of":18,"metrics":{"LPIPS":"0.162","PSNR":"28.69","SSIM":"0.9029"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-cbsd68-sigma50","task":"Color Image Denoising","dataset":"CBSD68 sigma50","model":"DMID-d","rank_in_archive_order":4,"of":18,"metrics":{"PSNR":"28.69"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-cbsd68-sigma50","task":"Color Image Denoising","dataset":"CBSD68 sigma50","model":"DMID-p","rank_in_archive_order":16,"of":18,"metrics":{"LPIPS":"0.122","PSNR":"26.63","SSIM":"0.8605"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-imagenet-sigma100","task":"Color Image Denoising","dataset":"ImageNet sigma100","model":"DMID-p","rank_in_archive_order":1,"of":2,"metrics":{"LPIPS":"0.156","PSNR":"24.61","SSIM":"0.7987"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-imagenet-sigma100","task":"Color Image Denoising","dataset":"ImageNet sigma100","model":"DMID-d","rank_in_archive_order":2,"of":2,"metrics":{"LPIPS":"0.201","PSNR":"27.00","SSIM":"0.8626"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-imagenet-sigma150","task":"Color Image Denoising","dataset":"ImageNet sigma150","model":"DMID-p","rank_in_archive_order":1,"of":2,"metrics":{"LPIPS":"0.259","PSNR":"22.94","SSIM":"0.6932"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-imagenet-sigma150","task":"Color Image Denoising","dataset":"ImageNet sigma150","model":"DMID-d","rank_in_archive_order":2,"of":2,"metrics":{"LPIPS":"0.295","PSNR":"25.39","SSIM":"0.8236"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-imagenet-sigma200","task":"Color Image Denoising","dataset":"ImageNet sigma200","model":"DMID-p","rank_in_archive_order":1,"of":2,"metrics":{"LPIPS":"0.259","PSNR":"21.56","SSIM":"0.6932"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-imagenet-sigma200","task":"Color Image Denoising","dataset":"ImageNet sigma200","model":"DMID-d","rank_in_archive_order":2,"of":2,"metrics":{"LPIPS":"0.295","PSNR":"24.25","SSIM":"0.7915"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-imagenet-sigma250","task":"Color Image Denoising","dataset":"ImageNet sigma250","model":"DMID-p","rank_in_archive_order":1,"of":2,"metrics":{"LPIPS":"0.289","PSNR":"20.87","SSIM":"0.6701"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-imagenet-sigma250","task":"Color Image Denoising","dataset":"ImageNet sigma250","model":"DMID-d","rank_in_archive_order":2,"of":2,"metrics":{"LPIPS":"0.346","PSNR":"23.44","SSIM":"0.7675"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-imagenet-sigma50","task":"Color Image Denoising","dataset":"ImageNet sigma50","model":"DMID-p","rank_in_archive_order":1,"of":2,"metrics":{"LPIPS":"0.087","PSNR":"27.59","SSIM":"0.8722"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-imagenet-sigma50","task":"Color Image Denoising","dataset":"ImageNet sigma50","model":"DMID-d","rank_in_archive_order":2,"of":2,"metrics":{"LPIPS":"0.114","PSNR":"29.9","SSIM":"0.9157"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-kodak-sigma50","task":"Color Image Denoising","dataset":"Kodak sigma50","model":"DMID-d","rank_in_archive_order":1,"of":1,"metrics":{"PSNR":"30.14"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-kodak24-sigma100","task":"Color Image Denoising","dataset":"Kodak24 sigma100","model":"DMID-p","rank_in_archive_order":1,"of":2,"metrics":{"LPIPS":"0.211","PSNR":"25.31","SSIM":"0.8107"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-kodak24-sigma100","task":"Color Image Denoising","dataset":"Kodak24 sigma100","model":"DMID-d","rank_in_archive_order":2,"of":2,"metrics":{"LPIPS":"0.274","PSNR":"27.5","SSIM":"0.8682"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-kodak24-sigma15","task":"Color Image Denoising","dataset":"Kodak24 sigma15","model":"DMID-d","rank_in_archive_order":1,"of":2,"metrics":{"PSNR":"35.51"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-kodak24-sigma150","task":"Color Image Denoising","dataset":"Kodak24 sigma150","model":"DMID-p","rank_in_archive_order":1,"of":2,"metrics":{"LPIPS":"0.211","PSNR":"24.06","SSIM":"0.7707"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-kodak24-sigma150","task":"Color Image Denoising","dataset":"Kodak24 sigma150","model":"DMID-d","rank_in_archive_order":2,"of":2,"metrics":{"LPIPS":"0.348","PSNR":"26.08","SSIM":"0.8336"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-kodak24-sigma200","task":"Color Image Denoising","dataset":"Kodak24 sigma200","model":"DMID-p","rank_in_archive_order":1,"of":2,"metrics":{"LPIPS":"0.318","PSNR":"22.99","SSIM":"0.7311"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-kodak24-sigma200","task":"Color Image Denoising","dataset":"Kodak24 sigma200","model":"DMID-d","rank_in_archive_order":2,"of":2,"metrics":{"LPIPS":"0.382","PSNR":"25.08","SSIM":"0.8054"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-kodak24-sigma25","task":"Color Image Denoising","dataset":"Kodak24 sigma25","model":"DMID-d","rank_in_archive_order":1,"of":3,"metrics":{"PSNR":"33.12"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-kodak24-sigma250","task":"Color Image Denoising","dataset":"Kodak24 sigma250","model":"DMID-p","rank_in_archive_order":1,"of":2,"metrics":{"LPIPS":"0.356","PSNR":"22.44","SSIM":"0.7133"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-kodak24-sigma250","task":"Color Image Denoising","dataset":"Kodak24 sigma250","model":"DMID-d","rank_in_archive_order":2,"of":2,"metrics":{"LPIPS":"0.446","PSNR":"24.4","SSIM":"0.7853"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-kodak24-sigma50","task":"Color Image Denoising","dataset":"Kodak24 sigma50","model":"DMID-d(MMSE)","rank_in_archive_order":1,"of":9,"metrics":{"LPIPS":"0.172","PSNR":"30.13","SSIM":"0.9174"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-kodak24-sigma50","task":"Color Image Denoising","dataset":"Kodak24 sigma50","model":"DMID-p","rank_in_archive_order":9,"of":9,"metrics":{"LPIPS":"0.131","PSNR":"27.9","SSIM":"0.8770"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-mcmaster-sigma100","task":"Color Image Denoising","dataset":"McMaster sigma100","model":"DMID-p","rank_in_archive_order":1,"of":2,"metrics":{"LPIPS":"0.158","PSNR":"25.37","SSIM":"0.8560"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-mcmaster-sigma100","task":"Color Image Denoising","dataset":"McMaster sigma100","model":"DMID-d","rank_in_archive_order":2,"of":2,"metrics":{"LPIPS":"0.206","PSNR":"27.57","SSIM":"0.8990"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-mcmaster-sigma15","task":"Color Image Denoising","dataset":"McMaster sigma15","model":"DMID-d","rank_in_archive_order":2,"of":4,"metrics":{"PSNR":"35.72"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-mcmaster-sigma150","task":"Color Image Denoising","dataset":"McMaster sigma150","model":"DMID-p","rank_in_archive_order":1,"of":2,"metrics":{"LPIPS":"0.209","PSNR":"23.72","SSIM":"0.8148"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-mcmaster-sigma150","task":"Color Image Denoising","dataset":"McMaster sigma150","model":"DMID-d","rank_in_archive_order":2,"of":2,"metrics":{"LPIPS":"0.267","PSNR":"25.89","SSIM":"0.8672"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-mcmaster-sigma200","task":"Color Image Denoising","dataset":"McMaster sigma200","model":"DMID-p","rank_in_archive_order":1,"of":2,"metrics":{"LPIPS":"0.252","PSNR":"22.51","SSIM":"0.7816"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-mcmaster-sigma200","task":"Color Image Denoising","dataset":"McMaster sigma200","model":"DMID-d","rank_in_archive_order":2,"of":2,"metrics":{"LPIPS":"0.303","PSNR":"24.68","SSIM":"0.8392"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-mcmaster-sigma25","task":"Color Image Denoising","dataset":"McMaster sigma25","model":"DMID-d","rank_in_archive_order":2,"of":4,"metrics":{"PSNR":"33.49"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-mcmaster-sigma250","task":"Color Image Denoising","dataset":"McMaster sigma250","model":"DMID-p","rank_in_archive_order":1,"of":2,"metrics":{"LPIPS":"0.290","PSNR":"21.73","SSIM":"0.7553"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-mcmaster-sigma250","task":"Color Image Denoising","dataset":"McMaster sigma250","model":"DMID-d","rank_in_archive_order":2,"of":2,"metrics":{"LPIPS":"0.359","PSNR":"23.81","SSIM":"0.8174"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-mcmaster-sigma50","task":"Color Image Denoising","dataset":"McMaster sigma50","model":"DMID-d","rank_in_archive_order":2,"of":7,"metrics":{"PSNR":"30.51"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-mcmaster-sigma50","task":"Color Image Denoising","dataset":"McMaster sigma50","model":"DMID-p","rank_in_archive_order":6,"of":7,"metrics":{"LPIPS":"0.092","PSNR":"28.34","SSIM":"0.9112"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-mcmaster-sigma50","task":"Color Image Denoising","dataset":"McMaster sigma50","model":"DMID-d(MMSE)","rank_in_archive_order":7,"of":7,"metrics":{"LPIPS":"0.124","SSIM":"0.9396"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-urban100-sigma25","task":"Color Image Denoising","dataset":"Urban100 sigma25","model":"DMID-d","rank_in_archive_order":3,"of":6,"metrics":{"PSNR":"33.11"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-urban100-sigma50","task":"Color Image Denoising","dataset":"Urban100 sigma50","model":"DMID-d","rank_in_archive_order":2,"of":9,"metrics":{"PSNR":"30.28"},"uses_additional_data":false},{"leaderboard":"/sota/color-image-denoising-on-urban100-sigma15-1","task":"Color Image Denoising","dataset":"urban100 sigma15","model":"DMID-d","rank_in_archive_order":2,"of":8,"metrics":{"Average PSNR":"35.26","PSNR":"35.26"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2307.03992","atlas_url":"https://app.syntology.ai/?focus=2307.03992","mcp":null,"developers":"https://syntology.ai/developers","held":true,"reason":"arxiv_id_in_atlas","checked_against":{"atlas_ids":549449,"atlas_ids_sha256":"afd72399dafd9906e9062927f70655952385a3d75488558e1f0484e950ab09b7","read_at":"2026-10-05T01:46:21Z"},"agent_calls":[{"tool":"get_citation_path","arguments":{"paper_1":"2307.03992"},"arguments_in":{"paper_2":"another paper's arXiv id or title"},"call":"get_citation_path(paper_1=\"2307.03992\", paper_2=\"…\")"},{"tool":"get_concepts_for_paper","arguments":{"arxiv_id":"2307.03992"},"call":"get_concepts_for_paper(arxiv_id=\"2307.03992\")"}],"by_repo":[{"repo":"li-tong-621/dmid","kind":"official","samples":0,"ran":0,"instrument":0,"unverified":0}],"by_repo_is":"one row per repository in the archive's code links, Syntology's graph links or the samples; samples 0 means no sample is linked to this paper, a statement about Syntology's coverage, not about the repository: with harvested_for_other_papers present, Syntology harvested the repository for that many other papers, and without it, Syntology harvested nothing from it; placed_by_identical_code counts samples whose link records no repository, placed here because identical code was harvested from this repository for another paper; ran counts executions on a synthesized input and instrument counts failures of Syntology's instrument, not of the code"},"code_links_note":null,"arxiv_metadata":null,"syntology_extracted_results":null,"record_sha256":"9b61e3a197bf080329123475b33b05cb6040d5ba052a34482d3be9b390280875","record_changed_at":"2026-09-28","record_changed_at_basis":"first_hashed"}