{"corpus_id":19328533,"paper_sha":"bdf81dcd76fa96e980e24cdd1d912b772b801f44","doi":"10.1109/CVPRW.2017.233","arxiv_id":"1706.01788","pmid":null,"pmcid":null,"mag_id":2621724949,"dblp_id":"journals/corr/AmeriniUBC17","acl_id":null,"title":"Localization of JPEG Double Compression Through Multi-domain Convolutional Neural Networks","year":2017,"publication_date":"2017-06-06","venue":"2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)","journal":{"name":"2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)","pages":"1865-1871","volume":null},"journal_issn":null,"journal_title":null,"publication_types":["JournalArticle","Conference"],"pubmed_pub_types":null,"s2_fields_of_study":["Computer Science"],"reference_count":25,"citation_count":136,"influential_citation_count":13,"is_open_access":true,"arxiv_categories":["cs.MM","cs.CR"],"arxiv_license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","arxiv_journal_ref":null,"mesh_headings":null,"chemicals":null,"comments_corrections":null,"source_flags":1,"s2_open_access_pdf_url":"http://arxiv.org/pdf/1706.01788","s2_open_access_landing_url":"https://www.semanticscholar.org/paper/bdf81dcd76fa96e980e24cdd1d912b772b801f44","s2_open_access_license":null,"s2_open_access_status":"GREEN","pmc_open_access_pdf_url":null,"pmc_open_access_landing_url":null,"pmc_open_access_license":null,"pmc_open_access_status":null,"unpaywall_open_access_pdf_url":null,"unpaywall_open_access_landing_url":null,"unpaywall_open_access_license":null,"unpaywall_open_access_status":null,"abstract":"When an attacker wants to falsify an image, in most of cases she/he will perform a JPEG recompression. Different techniques have been developed based on diverse theoretical assumptions but very effective solutions have not been developed yet. Recently, machine learning based approaches have been started to appear in the field of image forensics to solve diverse tasks such as acquisition source identification and forgery detection. In this last case, the aim ahead would be to get a trained neural network able, given a to-be-checked image, to reliably localize the forged areas. With this in mind, our paper proposes a step forward in this direction by analyzing how a single or double JPEG compression can be revealed and localized using convolutional neural networks (CNNs). Different kinds of input to the CNN have been taken into consideration, and various experiments have been carried out trying also to evidence potential issues to be further investigated.","claims":[{"public_id":"cl_c339f57d4090ee8114db217762a85941","status":"active","text":"Different kinds of CNN input were evaluated to study how compression artifacts can be detected and localized.","confidence":0.85,"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/claims/cl_c339f57d4090ee8114db217762a85941"},{"public_id":"cl_eaea4545e8be70189d9d1161b1e2c34a","status":"active","text":"Experiments identified potential issues that warrant further investigation in JPEG double compression localization.","confidence":0.75,"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/claims/cl_eaea4545e8be70189d9d1161b1e2c34a"},{"public_id":"cl_b233ab5d5d632c96628d608fc1c94e8d","status":"active","text":"Single or double JPEG compression can be revealed and localized using convolutional neural networks.","confidence":0.95,"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/claims/cl_b233ab5d5d632c96628d608fc1c94e8d"}],"concepts":[{"public_id":"co_26239aba2dc18dbc44234db96ccb7c76","status":"active","name":"CNN input","description":"The input representation or feature format provided to a convolutional neural network.","types":["model input"],"aliases":["input to the CNN"],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_26239aba2dc18dbc44234db96ccb7c76"},{"public_id":"co_5ff8d427b281815d934d4ed0b2d5aa9e","status":"active","name":"experiments","description":"Controlled empirical tests used to evaluate the proposed localization approach.","types":["evaluation"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_5ff8d427b281815d934d4ed0b2d5aa9e"},{"public_id":"co_7161b80e223dc87d44d5f6c47f3a65dc","status":"active","name":"forgery localization","description":"The task of identifying and localizing manipulated regions within an image.","types":["task"],"aliases":["localize the forged areas"],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_7161b80e223dc87d44d5f6c47f3a65dc"},{"public_id":"co_8f52703db8e66819dcbd5605491086d2","status":"active","name":"JPEG double compression","description":"The repeated compression of an image using JPEG, often studied for forensic detection.","types":["forensic artifact"],"aliases":["double JPEG compression"],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_8f52703db8e66819dcbd5605491086d2"},{"public_id":"co_99afa46194cd99388df4b27fddb7af9c","status":"active","name":"single JPEG compression","description":"An image that has been compressed once using the JPEG format.","types":["image processing artifact"],"aliases":["single compression"],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_99afa46194cd99388df4b27fddb7af9c"},{"public_id":"co_d2e41548a37306728d62a3bcf76f701f","status":"active","name":"double JPEG compression","description":"An image that has undergone JPEG compression twice.","types":["image processing artifact"],"aliases":["JPEG double compression"],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_d2e41548a37306728d62a3bcf76f701f"},{"public_id":"co_dd76741de1beec87868ca4081d57e6d0","status":"active","name":"convolutional neural networks","description":"Neural network models that use convolutional layers to learn spatial patterns from image data.","types":["method"],"aliases":["CNNs","CNN"],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_dd76741de1beec87868ca4081d57e6d0"},{"public_id":"co_f036ef80c812aeb2ac3fce8dc4a6ffc7","status":"active","name":"compression artifacts","description":"Visual or statistical traces introduced by image compression that can be used for analysis.","types":["phenomenon"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_f036ef80c812aeb2ac3fce8dc4a6ffc7"}],"external_ids":{"DOI":"10.1109/CVPRW.2017.233","ArXiv":"1706.01788","PubMed":null,"PubMedCentral":null,"MAG":2621724949,"DBLP":"journals/corr/AmeriniUBC17","ACL":null},"open_access":{"is_open_access":true,"pdf_url":"http://arxiv.org/pdf/1706.01788","landing_url":"https://www.semanticscholar.org/paper/bdf81dcd76fa96e980e24cdd1d912b772b801f44","source":"semantic_scholar","pdf_url_source":"semantic_scholar_open_access_pdf","license":null,"status":"GREEN","reason":null},"reference_availability":{"status":"available","references_indexed":true,"full_text_available":true,"full_text_source":"arxiv","count_basis":"semantic_scholar_metadata","extraction_status":"not_applicable","reason":null},"source":{"provider":"episteme2","base_corpus":"semantic_scholar_dump","freshness_mode":"unknown","basis":["semantic_scholar_metadata","postgres_metadata"],"limits":["paper metadata is based on indexed upstream scholarly datasets","claims and concepts are available only for extracted papers","absence of claims or concepts means no extracted graph data is available in this response"],"status":"available","degraded":false,"degraded_reasons":[],"diagnostics":{"status":"available","degraded":false,"degraded_reasons":[],"metadata_status":"available","graph_status":"available","abstract_status":"available"},"source_flags":1},"paper_id":636913,"paper_uid":"02e22405-4091-4fd2-9bdf-a32e0e2889ca","canonical_identity":{"paper_id":636913,"paper_uid":"02e22405-4091-4fd2-9bdf-a32e0e2889ca","identity_status":"available","lookup_basis":"semantic_scholar_external_id","compatibility_path":"corpus_id"},"url":"https://sah.borca.ai/papers/19328533"}