{"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/foundir-unleashing-million-scale-training","title":"FoundIR: Unleashing Million-scale Training Data to Advance Foundation Models for Image Restoration","arxiv_id":"2412.01427","date":"2024-12-02","proceeding":null,"authors":["Hao Li","Xiang Chen","Jiangxin Dong","Jinhui Tang","Jinshan Pan"],"abstract":"Despite the significant progress made by all-in-one models in universal image restoration, existing methods suffer from a generalization bottleneck in real-world scenarios, as they are mostly trained on small-scale synthetic datasets with limited degradations. Therefore, large-scale high-quality real-world training data is urgently needed to facilitate the emergence of foundational models for image restoration. To advance this field, we spare no effort in contributing a million-scale dataset with two notable advantages over existing training data: real-world samples with larger-scale, and degradation types with higher diversity. By adjusting internal camera settings and external imaging conditions, we can capture aligned image pairs using our well-designed data acquisition system over multiple rounds and our data alignment criterion. Moreover, we propose a robust model, FoundIR, to better address a broader range of restoration tasks in real-world scenarios, taking a further step toward foundation models. Specifically, we first utilize a diffusion-based generalist model to remove degradations by learning the degradation-agnostic common representations from diverse inputs, where incremental learning strategy is adopted to better guide model training. To refine the model's restoration capability in complex scenarios, we introduce degradation-aware specialist models for achieving final high-quality results. Extensive experiments show the value of our dataset and the effectiveness of our method.","url_abs":"https://arxiv.org/abs/2412.01427v1","url_pdf":"https://arxiv.org/pdf/2412.01427v1.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":"foundir-unleashing-million-scale-training","repo_url":"https://github.com/House-Leo/FoundIR","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"image-restoration","task_name":"Image Restoration"},{"task_slug":"incremental-learning","task_name":"Incremental Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2412.01427","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2412.01427"}},"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/House-Leo/FoundIR","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran":10,"ran_violates":2,"ran_draft_wrong":2,"ran_fixture":2,"unverified":8},"by_repo_kind":{"listed":{"samples":24,"ran":16,"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":"1d7ce6d6dd7ff5c1","entry":"calculate_fid","repo":"House-Leo/FoundIR","repo_kind":"listed","path":"metrics/fid.py","file_url":"https://github.com/House-Leo/FoundIR/blob/HEAD/metrics/fid.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":"1d7ce6d6dd7ff5c1"}},{"code_sha256_prefix":"a0a9dee5557b1557","entry":"calculate_weights_indices","repo":"House-Leo/FoundIR","repo_kind":"listed","path":"matlab_functions.py","file_url":"https://github.com/House-Leo/FoundIR/blob/HEAD/matlab_functions.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":"a0a9dee5557b1557"}},{"code_sha256_prefix":"5170ff1c4106b27a","entry":"compute_feature","repo":"House-Leo/FoundIR","repo_kind":"listed","path":"metrics/niqe.py","file_url":"https://github.com/House-Leo/FoundIR/blob/HEAD/metrics/niqe.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":"5170ff1c4106b27a"}},{"code_sha256_prefix":"2a78a7db11a55ede","entry":"cubic","repo":"House-Leo/FoundIR","repo_kind":"listed","path":"matlab_functions.py","file_url":"https://github.com/House-Leo/FoundIR/blob/HEAD/matlab_functions.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"2a78a7db11a55ede"}},{"code_sha256_prefix":"d1ef6b8cb9a28a53","entry":"default","repo":"House-Leo/FoundIR","repo_kind":"listed","path":"src/model.py","file_url":"https://github.com/House-Leo/FoundIR/blob/HEAD/src/model.py","link_basis":"harvester_set","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"d1ef6b8cb9a28a53"}},{"code_sha256_prefix":"a1282a48b0941f5d","entry":"estimate_aggd_param","repo":"House-Leo/FoundIR","repo_kind":"listed","path":"metrics/niqe.py","file_url":"https://github.com/House-Leo/FoundIR/blob/HEAD/metrics/niqe.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":"a1282a48b0941f5d"}},{"code_sha256_prefix":"608e364a9d2376a3","entry":"exists","repo":"House-Leo/FoundIR","repo_kind":"listed","path":"src/model.py","file_url":"https://github.com/House-Leo/FoundIR/blob/HEAD/src/model.py","link_basis":"harvester_set","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"608e364a9d2376a3"}},{"code_sha256_prefix":"c9fb213c957dc82d","entry":"extract_inception_features","repo":"House-Leo/FoundIR","repo_kind":"listed","path":"metrics/fid.py","file_url":"https://github.com/House-Leo/FoundIR/blob/HEAD/metrics/fid.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":"c9fb213c957dc82d"}},{"code_sha256_prefix":"ef9faf68f492a375","entry":"flow_warp","repo":"House-Leo/FoundIR","repo_kind":"listed","path":"specialist_model/basicsr/archs/arch_util.py","file_url":"https://github.com/House-Leo/FoundIR/blob/HEAD/specialist_model/basicsr/archs/arch_util.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"ef9faf68f492a375"}},{"code_sha256_prefix":"96ad5dc9ca239aec","entry":"make_layer","repo":"House-Leo/FoundIR","repo_kind":"listed","path":"specialist_model/basicsr/archs/arch_util.py","file_url":"https://github.com/House-Leo/FoundIR/blob/HEAD/specialist_model/basicsr/archs/arch_util.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"96ad5dc9ca239aec"}},{"code_sha256_prefix":"95c6eba9350a455a","entry":"prepare_for_ssim","repo":"House-Leo/FoundIR","repo_kind":"listed","path":"metrics/psnr_ssim.py","file_url":"https://github.com/House-Leo/FoundIR/blob/HEAD/metrics/psnr_ssim.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"95c6eba9350a455a"}},{"code_sha256_prefix":"fb93712db7064ac4","entry":"prepare_for_ssim_rgb","repo":"House-Leo/FoundIR","repo_kind":"listed","path":"metrics/psnr_ssim.py","file_url":"https://github.com/House-Leo/FoundIR/blob/HEAD/metrics/psnr_ssim.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"fb93712db7064ac4"}},{"code_sha256_prefix":"95067518dc16b3e5","entry":"reorder_image","repo":"House-Leo/FoundIR","repo_kind":"listed","path":"metrics/metric_util.py","file_url":"https://github.com/House-Leo/FoundIR/blob/HEAD/metrics/metric_util.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":"95067518dc16b3e5"}},{"code_sha256_prefix":"5a1d8458dc7077a6","entry":"resize_flow","repo":"House-Leo/FoundIR","repo_kind":"listed","path":"specialist_model/basicsr/archs/arch_util.py","file_url":"https://github.com/House-Leo/FoundIR/blob/HEAD/specialist_model/basicsr/archs/arch_util.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":"5a1d8458dc7077a6"}},{"code_sha256_prefix":"82a15cc1e46f7e4d","entry":"to_3d","repo":"House-Leo/FoundIR","repo_kind":"listed","path":"specialist_model/basicsr/archs/Restormer_arch.py","file_url":"https://github.com/House-Leo/FoundIR/blob/HEAD/specialist_model/basicsr/archs/Restormer_arch.py","link_basis":"harvester_set","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"82a15cc1e46f7e4d"}},{"code_sha256_prefix":"b20f2a5df739a59e","entry":"to_4d","repo":"House-Leo/FoundIR","repo_kind":"listed","path":"specialist_model/basicsr/archs/Restormer_arch.py","file_url":"https://github.com/House-Leo/FoundIR/blob/HEAD/specialist_model/basicsr/archs/Restormer_arch.py","link_basis":"harvester_set","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"b20f2a5df739a59e"}},{"code_sha256_prefix":"7ab32da4188a2275","entry":"check_image_size","repo":"House-Leo/FoundIR","repo_kind":"listed","path":"specialist_model/inference_lowlight.py","file_url":"https://github.com/House-Leo/FoundIR/blob/HEAD/specialist_model/inference_lowlight.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":"7ab32da4188a2275"}},{"code_sha256_prefix":"3d48eb1a78a327d1","entry":"check_image_size","repo":"House-Leo/FoundIR","repo_kind":"listed","path":"specialist_model/inference_weather.py","file_url":"https://github.com/House-Leo/FoundIR/blob/HEAD/specialist_model/inference_weather.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":"3d48eb1a78a327d1"}},{"code_sha256_prefix":"aca6d1049c83a507","entry":"dict2str","repo":"House-Leo/FoundIR","repo_kind":"listed","path":"cal_metrics.py","file_url":"https://github.com/House-Leo/FoundIR/blob/HEAD/cal_metrics.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":"aca6d1049c83a507"}},{"code_sha256_prefix":"101e90f511df891e","entry":"img2tensor","repo":"House-Leo/FoundIR","repo_kind":"listed","path":"cal_metrics.py","file_url":"https://github.com/House-Leo/FoundIR/blob/HEAD/cal_metrics.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":"101e90f511df891e"}},{"code_sha256_prefix":"dfb4171d25811492","entry":"imresize","repo":"House-Leo/FoundIR","repo_kind":"listed","path":"matlab_functions.py","file_url":"https://github.com/House-Leo/FoundIR/blob/HEAD/matlab_functions.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":"dfb4171d25811492"}},{"code_sha256_prefix":"0029cb1566e47350","entry":"niqe","repo":"House-Leo/FoundIR","repo_kind":"listed","path":"metrics/niqe.py","file_url":"https://github.com/House-Leo/FoundIR/blob/HEAD/metrics/niqe.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":"0029cb1566e47350"}},{"code_sha256_prefix":"7952dbdafdde6e7a","entry":"scandir","repo":"House-Leo/FoundIR","repo_kind":"listed","path":"cal_metrics.py","file_url":"https://github.com/House-Leo/FoundIR/blob/HEAD/cal_metrics.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":"7952dbdafdde6e7a"}},{"code_sha256_prefix":"ac6cfb6f7e2c39d2","entry":"tensor2img","repo":"House-Leo/FoundIR","repo_kind":"listed","path":"src/model.py","file_url":"https://github.com/House-Leo/FoundIR/blob/HEAD/src/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":"ac6cfb6f7e2c39d2"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}