{"url":"/dataset/icb","name":"ICB","full_name":"Image Compression Benchmark","description_markdown":"A carefully chosen set of high-resolution high-precision natural images suited for compression algorithm evaluation.\r\n\r\nThe images historically used for compression research (lena, barbra, pepper etc...) have outlived their useful life and its about time they become a part of history only. They are too small, come from data sources too old and are available in only 8-bit precision.\r\n\r\nThese high-resolution high-precision images have been carefully selected to aid in image compression research and algorithm evaluation. These are photographic images chosen to come from a wide variety of sources and each one picked to stress different aspects of algorithms. Images are available in 8-bit, 16-bit and 16-bit linear variations, RGB and gray.\r\n\r\nThese Images are available without any prohibitive copyright restrictions.\r\n\r\nThese images are (c) there respective owners. You are granted full redistribution and publication rights on these images provided:\r\n\r\n1. The origin of the pictures must not be misrepresented; you must not claim that you took the original pictures. If you use, publish or redistribute them, an acknowledgment would be appreciated but is not required.\r\n2. Altered versions must be plainly marked as such, and must not be misinterpreted as being the originals.\r\n3. No payment is required for distribution of this material, it must be available freely under the conditions stated here. That is, it is prohibited to sell the material.\r\n4. This notice may not be removed or altered from any distribution.\r\n\r\n*For grayscale evaluation, use the Grayscale 8 bit dataset, for color evaluation, use the Color 8 bit dataset.*\r\n\r\n```\r\n@online{icb,\r\n  author = {Rawzor},\r\n  title  = {Image Compression Benchmark},\r\n  url    = {http://imagecompression.info/}\r\n}\r\n```","description_withheld":null,"homepage":"https://imagecompression.info/","introduced_date":null,"introduced_date_note":null,"introduced_by":null,"license":null,"modalities":[{"name":"Images","url":"/datasets/modality/images"}],"tasks":[{"name":"JPEG Artifact Correction","url":"/task/jpeg-artifact-correction","datasets_with_task":"/datasets/task/jpeg-artifact-correction"}],"languages":[],"variants":["ICB","ICB (Quality 10 Color)","ICB (Quality 10 Grayscale)","ICB (Quality 20 Color)","ICB (Quality 30 Color)","ICB (Quality 20 Grayscale)","ICB (Quality 30 Grayscale)"],"data_loaders":[],"num_papers_in_archive":7,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[{"leaderboard":"/sota/jpeg-artifact-correction-on-icb-quality-10","task":"JPEG Artifact Correction","dataset_variant":"ICB (Quality 10 Color)","rows":6,"metrics":["PSNR","PSNR-B","SSIM"],"first_row_in_archive_order":{"model":"FBCNN","paper":"/paper/towards-flexible-blind-jpeg-artifacts-removal","metrics":{"PSNR":"32.18","PSNR-B":"32.15","SSIM":"0.815"},"code_links":[{"title":"jiaxi-jiang/fbcnn","url":"https://github.com/jiaxi-jiang/fbcnn"},{"title":"olaviinha/NeuralImageSuperResolution","url":"https://github.com/olaviinha/NeuralImageSuperResolution"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/jpeg-artifact-correction-on-icb-quality-20","task":"JPEG Artifact Correction","dataset_variant":"ICB (Quality 20 Color)","rows":6,"metrics":["PSNR","PSNR-B","SSIM"],"first_row_in_archive_order":{"model":"FBCNN","paper":"/paper/towards-flexible-blind-jpeg-artifacts-removal","metrics":{"PSNR":"34.38","PSNR-B":"34.34","SSIM":"0.844"},"code_links":[{"title":"jiaxi-jiang/fbcnn","url":"https://github.com/jiaxi-jiang/fbcnn"},{"title":"olaviinha/NeuralImageSuperResolution","url":"https://github.com/olaviinha/NeuralImageSuperResolution"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/jpeg-artifact-correction-on-icb-quality-10-1","task":"JPEG Artifact Correction","dataset_variant":"ICB (Quality 10 Grayscale)","rows":5,"metrics":["PSNR","PSNR-B","SSIM"],"first_row_in_archive_order":{"model":"QGAC","paper":"/paper/quantization-guided-jpeg-artifact-correction","metrics":{"PSNR":"34.73","PSNR-B":"34.58","SSIM":"0.896"},"code_links":[{"title":"Queuecumber/quantization-guided-ac","url":"https://gitlab.com/Queuecumber/quantization-guided-ac"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/jpeg-artifact-correction-on-icb-quality-20-1","task":"JPEG Artifact Correction","dataset_variant":"ICB (Quality 20 Grayscale)","rows":5,"metrics":["PSNR","PSNR-B","SSIM"],"first_row_in_archive_order":{"model":"QGAC","paper":"/paper/quantization-guided-jpeg-artifact-correction","metrics":{"PSNR":"37.12","PSNR-B":"36.88","SSIM":"0.924"},"code_links":[{"title":"Queuecumber/quantization-guided-ac","url":"https://gitlab.com/Queuecumber/quantization-guided-ac"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/jpeg-artifact-correction-on-icb-quality-30","task":"JPEG Artifact Correction","dataset_variant":"ICB (Quality 30 Color)","rows":4,"metrics":["PSNR","PSNR-B","SSIM"],"first_row_in_archive_order":{"model":"FBCNN","paper":"/paper/towards-flexible-blind-jpeg-artifacts-removal","metrics":{"PSNR":"35.41","PSNR-B":"35.35","SSIM":"0.857"},"code_links":[{"title":"jiaxi-jiang/fbcnn","url":"https://github.com/jiaxi-jiang/fbcnn"},{"title":"olaviinha/NeuralImageSuperResolution","url":"https://github.com/olaviinha/NeuralImageSuperResolution"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/jpeg-artifact-correction-on-icb-quality-30-1","task":"JPEG Artifact Correction","dataset_variant":"ICB (Quality 30 Grayscale)","rows":1,"metrics":["PSNR"],"first_row_in_archive_order":{"model":"QGAC","paper":"/paper/quantization-guided-jpeg-artifact-correction","metrics":{"PSNR":"38.43"},"code_links":[{"title":"Queuecumber/quantization-guided-ac","url":"https://gitlab.com/Queuecumber/quantization-guided-ac"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"}],"papers_with_a_benchmark_row":[{"paper":"/paper/towards-flexible-blind-jpeg-artifacts-removal","title":"Towards Flexible Blind JPEG Artifacts Removal","date":"2021-09-29","rows_on_this_dataset":3,"code_links":2,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":11,"samples_ran":9,"samples_unverified":2,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/quantization-guided-jpeg-artifact-correction","title":"Quantization Guided JPEG Artifact Correction","date":"2020-04-17","rows_on_this_dataset":6,"code_links":1,"syntology":null},{"paper":"/paper/implicit-dual-domain-convolutional-network","title":"Implicit Dual-domain Convolutional Network for Robust Color Image Compression Artifact Reduction","date":"2018-10-18","rows_on_this_dataset":4,"code_links":0,"syntology":null},{"paper":"/paper/dmcnn-dual-domain-multi-scale-convolutional","title":"DMCNN: Dual-Domain Multi-Scale Convolutional Neural Network for Compression Artifacts Removal","date":"2018-06-08","rows_on_this_dataset":4,"code_links":0,"syntology":null},{"paper":"/paper/multi-level-wavelet-cnn-for-image-restoration","title":"Multi-level Wavelet-CNN for Image Restoration","date":"2018-05-18","rows_on_this_dataset":5,"code_links":5,"syntology":null},{"paper":"/paper/compression-artifacts-reduction-by-a-deep","title":"Compression Artifacts Reduction by a Deep Convolutional Network","date":"2015-04-27","rows_on_this_dataset":5,"code_links":4,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":12,"samples_ran":0,"samples_unverified":12,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}}],"syntology_totals":{"read_at":"2026-09-24T18:15:14+00:00","papers_with_samples":2,"samples_harvested":23,"samples_ran":9,"samples_unverified":14,"pointer_only_for_licence":0,"papers_with_no_sample_that_ran":1,"note":"the per-paper counts above, summed; not a rate"},"papers_note":"The archive never published its papers-using-dataset list; these are papers with a leaderboard row on this dataset's benchmarks."}