{"url":"/task/image-compression","name":"Image Compression","slug":"image-compression","description_markdown":"**Image Compression** is an application of data compression for digital images to lower their storage and/or transmission requirements.\r\n\r\n\r\n<span class=\"description-source\">Source: [Variable Rate Deep Image Compression With a Conditional Autoencoder ](https://arxiv.org/abs/1909.04802)</span>","categories":[{"name":"Computer Vision","url":"/area/computer-vision"}],"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","slug_source":"archive_url"},"counts":{"papers_tagged":1008,"papers_with_code":320,"benchmarks":11,"benchmark_tables_in_archive":11,"benchmark_tables_shown":11,"benchmark_tables_withheld_as_spam":0,"benchmark_definition":"a leaderboard table with at least one row; benchmark_tables_shown also counts the zero-row tables; benchmark_tables_in_archive adds the tables withheld as spam","datasets":12,"subtasks":4,"parent_tasks":0},"benchmarks":[{"leaderboard":"/sota/image-compression-on-kodak","slug":"image-compression-on-kodak","dataset":"kodak","dataset_url":null,"rows_in_archive":8,"metrics":["BD-Rate over VTM-17.0"],"first_row_in_archive_order":{"model":"MLIC+","paper_title":"MLIC: Multi-Reference Entropy Model for Learned Image Compression","paper_url":"/paper/multi-reference-entropy-model-for-learned","paper_date":"2022-11-14","arxiv_id":"2211.07273","code_links":[{"title":"jiangweibeta/mlic","url":"https://github.com/jiangweibeta/mlic"}],"syntology":null}},{"leaderboard":"/sota/image-compression-on-imagenet32","slug":"image-compression-on-imagenet32","dataset":"ImageNet32","dataset_url":"/dataset/imagenet-32","rows_in_archive":5,"metrics":["bpsp"],"first_row_in_archive_order":{"model":"iFlow","paper_title":"iFlow: Numerically Invertible Flows for Efficient Lossless Compression via a Uniform Coder","paper_url":"/paper/iflow-numerically-invertible-flows-for","paper_date":"2021-11-01","arxiv_id":"2111.00965","code_links":[],"syntology":null}},{"leaderboard":"/sota/image-compression-on-bsds500","slug":"image-compression-on-bsds500","dataset":"BSDS500","dataset_url":"/dataset/bsds500","rows_in_archive":1,"metrics":["Average PSNR","SSIM"],"first_row_in_archive_order":{"model":"RK-CCSNet","paper_title":"Sequential Convolution and Runge-Kutta Residual Architecture for Image Compressed Sensing","paper_url":"/paper/sequential-convolution-and-runge-kutta","paper_date":"","arxiv_id":null,"code_links":[{"title":"rkteddy/RK-CCSNet","url":"https://github.com/rkteddy/RK-CCSNet"}],"syntology":null}},{"leaderboard":"/sota/image-compression-on-caltech101","slug":"image-compression-on-caltech101","dataset":"Caltech101","dataset_url":null,"rows_in_archive":1,"metrics":["Bit rate"],"first_row_in_archive_order":{"model":"Lossyless Compressor","paper_title":"Lossy Compression for Lossless Prediction","paper_url":"/paper/lossy-compression-for-lossless-prediction","paper_date":"2021-06-21","arxiv_id":"2106.10800","code_links":[{"title":"YannDubs/lossyless","url":"https://github.com/YannDubs/lossyless"}],"syntology":{"n":11,"n_ran":2,"n_unverified":9,"n_pointer_only":0}}},{"leaderboard":"/sota/image-compression-on-cars-196","slug":"image-compression-on-cars-196","dataset":"Cars-196","dataset_url":null,"rows_in_archive":1,"metrics":["Bit rate"],"first_row_in_archive_order":{"model":"Lossyless Compressor","paper_title":"Lossy Compression for Lossless Prediction","paper_url":"/paper/lossy-compression-for-lossless-prediction","paper_date":"2021-06-21","arxiv_id":"2106.10800","code_links":[{"title":"YannDubs/lossyless","url":"https://github.com/YannDubs/lossyless"}],"syntology":{"n":11,"n_ran":2,"n_unverified":9,"n_pointer_only":0}}},{"leaderboard":"/sota/image-compression-on-cifar-10","slug":"image-compression-on-cifar-10","dataset":"CIFAR-10","dataset_url":"/dataset/cifar-10","rows_in_archive":1,"metrics":["Bit rate"],"first_row_in_archive_order":{"model":"Lossyless Compressor","paper_title":"Lossy Compression for Lossless Prediction","paper_url":"/paper/lossy-compression-for-lossless-prediction","paper_date":"2021-06-21","arxiv_id":"2106.10800","code_links":[{"title":"YannDubs/lossyless","url":"https://github.com/YannDubs/lossyless"}],"syntology":{"n":11,"n_ran":2,"n_unverified":9,"n_pointer_only":0}}},{"leaderboard":"/sota/image-compression-on-fer2013","slug":"image-compression-on-fer2013","dataset":"FER2013","dataset_url":"/dataset/fer2013","rows_in_archive":1,"metrics":["10%"],"first_row_in_archive_order":{"model":"SUDHEER","paper_title":"Deep Convolutional Neural Network for Image Deconvolution","paper_url":"/paper/deep-convolutional-neural-network-for-image","paper_date":"2014-12-01","arxiv_id":null,"code_links":[],"syntology":null}},{"leaderboard":"/sota/image-compression-on-food-101","slug":"image-compression-on-food-101","dataset":"Food-101","dataset_url":"/dataset/food-101","rows_in_archive":1,"metrics":["Bit rate"],"first_row_in_archive_order":{"model":"Lossyless Compressor","paper_title":"Lossy Compression for Lossless Prediction","paper_url":"/paper/lossy-compression-for-lossless-prediction","paper_date":"2021-06-21","arxiv_id":"2106.10800","code_links":[{"title":"YannDubs/lossyless","url":"https://github.com/YannDubs/lossyless"}],"syntology":{"n":11,"n_ran":2,"n_unverified":9,"n_pointer_only":0}}},{"leaderboard":"/sota/image-compression-on-oxford-iiit-pets","slug":"image-compression-on-oxford-iiit-pets","dataset":"Oxford-IIIT Pet Dataset","dataset_url":"/dataset/oxford-iiit-pets","rows_in_archive":1,"metrics":["Bit rate"],"first_row_in_archive_order":{"model":"Lossyless Compressor","paper_title":"Lossy Compression for Lossless Prediction","paper_url":"/paper/lossy-compression-for-lossless-prediction","paper_date":"2021-06-21","arxiv_id":"2106.10800","code_links":[{"title":"YannDubs/lossyless","url":"https://github.com/YannDubs/lossyless"}],"syntology":{"n":11,"n_ran":2,"n_unverified":9,"n_pointer_only":0}}},{"leaderboard":"/sota/image-compression-on-pcam","slug":"image-compression-on-pcam","dataset":"PCam","dataset_url":"/dataset/pcam","rows_in_archive":1,"metrics":["Bit rate"],"first_row_in_archive_order":{"model":"Lossyless Compressor","paper_title":"Lossy Compression for Lossless Prediction","paper_url":"/paper/lossy-compression-for-lossless-prediction","paper_date":"2021-06-21","arxiv_id":"2106.10800","code_links":[{"title":"YannDubs/lossyless","url":"https://github.com/YannDubs/lossyless"}],"syntology":{"n":11,"n_ran":2,"n_unverified":9,"n_pointer_only":0}}},{"leaderboard":"/sota/image-compression-on-stl-10","slug":"image-compression-on-stl-10","dataset":"STL-10","dataset_url":"/dataset/stl-10","rows_in_archive":1,"metrics":["Bit rate"],"first_row_in_archive_order":{"model":"Lossyless Compressor","paper_title":"Lossy Compression for Lossless Prediction","paper_url":"/paper/lossy-compression-for-lossless-prediction","paper_date":"2021-06-21","arxiv_id":"2106.10800","code_links":[{"title":"YannDubs/lossyless","url":"https://github.com/YannDubs/lossyless"}],"syntology":{"n":11,"n_ran":2,"n_unverified":9,"n_pointer_only":0}}}],"datasets":[{"url":"/dataset/cifar-10","name":"CIFAR-10","full_name":"CIFAR-10","num_papers_in_archive":16145},{"url":"/dataset/stl-10","name":"STL-10","full_name":"Self-Taught Learning 10","num_papers_in_archive":1092},{"url":"/dataset/food-101","name":"Food-101","full_name":"","num_papers_in_archive":805},{"url":"/dataset/bsds500","name":"BSDS500","full_name":"Berkeley Segmentation Dataset 500","num_papers_in_archive":261},{"url":"/dataset/fer2013","name":"FER2013","full_name":"Facial Expression Recognition 2013 Dataset","num_papers_in_archive":168},{"url":"/dataset/imagenet-32","name":"ImageNet-32","full_name":"","num_papers_in_archive":112},{"url":"/dataset/pcam","name":"PCam","full_name":"PatchCamelyon","num_papers_in_archive":110},{"url":"/dataset/oxford-iiit-pets","name":"Oxford-IIIT Pet Dataset","full_name":"Oxford-IIIT Pet Dataset","num_papers_in_archive":90},{"url":"/dataset/oxford-iiit-pets-1","name":"Oxford-IIIT Pets","full_name":"","num_papers_in_archive":59},{"url":"/dataset/upiq","name":"UPIQ","full_name":"Unified Photometric Image Quality","num_papers_in_archive":7},{"url":"/dataset/osn-transmission-mini-celeba","name":"OSN-transmission_mini_CelebA","full_name":"","num_papers_in_archive":1},{"url":"/dataset/pinet","name":"PINet","full_name":"","num_papers_in_archive":0}],"subtasks":[{"url":"/task/color-image-compression-artifact-reduction","name":"Color Image Compression Artifact Reduction"},{"url":"/task/feature-compression","name":"Feature Compression"},{"url":"/task/jpeg-compression-artifact-reduction","name":"Jpeg Compression Artifact Reduction"},{"url":"/task/lossy-compression-artifact-reduction","name":"Lossy-Compression Artifact Reduction"}],"parent_tasks":[],"papers":{"order":"repositories listed in the archive (desc), then date (desc); the archive holds no stars","population":"papers tagged with this task that list at least one repository in the archive","shown":30,"of":320,"tagged_in_all":1008,"items":[{"url":"/paper/variational-image-compression-with-a-scale","title":"Variational image compression with a scale hyperprior","date":"2018-02-01","arxiv_id":"1802.01436","repositories_listed":16,"syntology":{"n":5,"n_ran":5,"n_unverified":0,"n_pointer_only":5}},{"url":"/paper/end-to-end-optimized-image-compression","title":"End-to-end Optimized Image Compression","date":"2016-11-05","arxiv_id":"1611.01704","repositories_listed":14,"syntology":{"n":20,"n_ran":4,"n_unverified":16,"n_pointer_only":4}},{"url":"/paper/zero-shot-super-resolution-using-deep","title":"\"Zero-Shot\" Super-Resolution using Deep Internal Learning","date":"2017-12-17","arxiv_id":"1712.06087","repositories_listed":7,"syntology":null},{"url":"/paper/full-resolution-image-compression-with","title":"Full Resolution Image Compression with Recurrent Neural Networks","date":"2016-08-18","arxiv_id":"1608.05148","repositories_listed":7,"syntology":{"n":1,"n_ran":0,"n_unverified":1,"n_pointer_only":0}},{"url":"/paper/qvrf-a-quantization-error-aware-variable-rate","title":"QVRF: A Quantization-error-aware Variable Rate Framework for Learned Image Compression","date":"2023-03-10","arxiv_id":"2303.05744","repositories_listed":6,"syntology":null},{"url":"/paper/elic-efficient-learned-image-compression-with","title":"ELIC: Efficient Learned Image Compression with Unevenly Grouped Space-Channel Contextual Adaptive Coding","date":"2022-03-21","arxiv_id":"2203.10886","repositories_listed":6,"syntology":{"n":12,"n_ran":8,"n_unverified":4,"n_pointer_only":0}},{"url":"/paper/an-end-to-end-compression-framework-based-on","title":"An End-to-End Compression Framework Based on Convolutional Neural Networks","date":"2017-08-02","arxiv_id":"1708.00838","repositories_listed":5,"syntology":null},{"url":"/paper/checkerboard-context-model-for-efficient","title":"Checkerboard Context Model for Efficient Learned Image Compression","date":"2021-03-29","arxiv_id":"2103.15306","repositories_listed":4,"syntology":{"n":2,"n_ran":2,"n_unverified":0,"n_pointer_only":0}},{"url":"/paper/compressai-a-pytorch-library-and-evaluation","title":"CompressAI: a PyTorch library and evaluation platform for end-to-end compression research","date":"2020-11-05","arxiv_id":"2011.03029","repositories_listed":4,"syntology":{"n":6,"n_ran":6,"n_unverified":0,"n_pointer_only":6}},{"url":"/paper/high-fidelity-generative-image-compression","title":"High-Fidelity Generative Image Compression","date":"2020-06-17","arxiv_id":"2006.09965","repositories_listed":4,"syntology":null},{"url":"/paper/learned-image-compression-with-discretized","title":"Learned Image Compression with Discretized Gaussian Mixture Likelihoods and Attention Modules","date":"2020-01-06","arxiv_id":"2001.01568","repositories_listed":4,"syntology":{"n":14,"n_ran":0,"n_unverified":14,"n_pointer_only":0}},{"url":"/paper/practical-full-resolution-learned-lossless","title":"Practical Full Resolution Learned Lossless Image Compression","date":"2018-11-30","arxiv_id":"1811.12817","repositories_listed":4,"syntology":null},{"url":"/paper/lossy-image-compression-with-compressive","title":"Lossy Image Compression with Compressive Autoencoders","date":"2017-03-01","arxiv_id":"1703.00395","repositories_listed":4,"syntology":{"n":1,"n_ran":0,"n_unverified":1,"n_pointer_only":0}},{"url":"/paper/learned-image-compression-with-mixed","title":"Learned Image Compression with Mixed Transformer-CNN Architectures","date":"2023-03-27","arxiv_id":"2303.14978","repositories_listed":3,"syntology":{"n":9,"n_ran":6,"n_unverified":3,"n_pointer_only":0}},{"url":"/paper/transformer-based-transform-coding","title":"Transformer-based Transform Coding","date":"2021-09-29","arxiv_id":null,"repositories_listed":3,"syntology":null},{"url":"/paper/deep-stereo-image-compression-with-decoder","title":"Neural Distributed Image Compression using Common Information","date":"2021-06-22","arxiv_id":"2106.11723","repositories_listed":3,"syntology":null},{"url":"/paper/m2tr-multi-modal-multi-scale-transformers-for","title":"M2TR: Multi-modal Multi-scale Transformers for Deepfake Detection","date":"2021-04-20","arxiv_id":"2104.09770","repositories_listed":3,"syntology":{"n":11,"n_ran":4,"n_unverified":7,"n_pointer_only":0}},{"url":"/paper/channel-wise-autoregressive-entropy-models","title":"Channel-wise Autoregressive Entropy Models for Learned Image Compression","date":"2020-07-17","arxiv_id":"2007.08739","repositories_listed":3,"syntology":null},{"url":"/paper/learning-for-video-compression-with","title":"Learning for Video Compression with Hierarchical Quality and Recurrent Enhancement","date":"2020-03-04","arxiv_id":"2003.01966","repositories_listed":3,"syntology":{"n":2,"n_ran":0,"n_unverified":2,"n_pointer_only":0}},{"url":"/paper/deep-image-compression-using-decoder-side","title":"Deep Image Compression using Decoder Side Information","date":"2020-01-14","arxiv_id":"2001.04753","repositories_listed":3,"syntology":null},{"url":"/paper/learning-to-compress-and-search-visual-data","title":"Learning to compress and search visual data in large-scale systems","date":"2019-01-24","arxiv_id":"1901.08437","repositories_listed":3,"syntology":null},{"url":"/paper/residual-dense-network-for-image-restoration","title":"Residual Dense Network for Image Restoration","date":"2018-12-25","arxiv_id":"1812.10477","repositories_listed":3,"syntology":null},{"url":"/paper/joint-autoregressive-and-hierarchical-priors","title":"Joint Autoregressive and Hierarchical Priors for Learned Image Compression","date":"2018-09-08","arxiv_id":"1809.02736","repositories_listed":3,"syntology":null},{"url":"/paper/efficient-nonlinear-transforms-for-lossy","title":"Efficient Nonlinear Transforms for Lossy Image Compression","date":"2018-01-31","arxiv_id":"1802.00847","repositories_listed":3,"syntology":null},{"url":"/paper/semantic-perceptual-image-compression-using","title":"Semantic Perceptual Image Compression using Deep Convolution Networks","date":"2016-12-27","arxiv_id":"1612.08712","repositories_listed":3,"syntology":null},{"url":"/paper/privacy-preserving-face-recognition-using-1","title":"Privacy-Preserving Face Recognition Using Trainable Feature Subtraction","date":"2024-03-19","arxiv_id":"2403.12457","repositories_listed":2,"syntology":null},{"url":"/paper/image-coding-for-machines-with-edge","title":"Image Coding for Machines with Edge Information Learning Using Segment Anything","date":"2024-03-07","arxiv_id":"2403.04173","repositories_listed":2,"syntology":null},{"url":"/paper/region-adaptive-transform-with-segmentation","title":"Region-Adaptive Transform with Segmentation Prior for Image Compression","date":"2024-03-01","arxiv_id":"2403.00628","repositories_listed":2,"syntology":{"n":18,"n_ran":14,"n_unverified":4,"n_pointer_only":18}},{"url":"/paper/misc-ultra-low-bitrate-image-semantic","title":"MISC: Ultra-low Bitrate Image Semantic Compression Driven by Large Multimodal Model","date":"2024-02-26","arxiv_id":"2402.16749","repositories_listed":2,"syntology":null},{"url":"/paper/compact-3d-scene-representation-via-self","title":"Compact 3D Scene Representation via Self-Organizing Gaussian Grids","date":"2023-12-19","arxiv_id":"2312.13299","repositories_listed":2,"syntology":{"n":1,"n_ran":1,"n_unverified":0,"n_pointer_only":1}}],"syntology_records":13,"syntology_note":"a paper without a record is not a recorded non-run: it may lack an arXiv id or simply be absent from the graph layer"},"description_links":{"kept":0,"unwrapped_to_text":0,"bare_urls_linked":0,"relative_images_dropped":0,"rule":"internal links are kept only when the target slug exists in the catalog"},"syntology":{"read_at":"2026-09-24T18:15:14+00:00","claim":"Per-sample execution status on synthesized fixtures ('ran N of M samples'); not a correctness claim and not a ranking signal.","status_vocabulary":{"ran_honours":"ran, honoured the contract we drafted","ran_violates":"ran, violated the contract we drafted","ran_draft_wrong":"ran; our contract draft was wrong, not the code","ran_fixture":"ran; our fixture could not drive it","ran":"ran on a synthesized input","unverified":"unverified (harvested, no recorded run)"}},"not_shown":{"libraries":"the archive has no per-task library table","trend_sparklines":"the Trend column of the benchmarks table was a rendered image; it is not in the archive","social_and_latest_sorts":"stars and social signals are not in the archive"}}