{"corpus_id":206764601,"paper_sha":"3ebc7ec2a9b89ebcc00328b11a1984d69a57f2a7","doi":"10.1109/TPAMI.2009.77","arxiv_id":null,"pmid":20299707,"pmcid":null,"mag_id":2104853049,"dblp_id":"journals/pami/TolaLF10","acl_id":null,"title":"DAISY: An Efficient Dense Descriptor Applied to Wide-Baseline Stereo","year":2010,"publication_date":"2010-05-01","venue":"IEEE Transactions on Pattern Analysis and Machine Intelligence","journal":{"name":"IEEE Transactions on Pattern Analysis and Machine Intelligence","pages":"815-830","volume":"32"},"journal_issn":null,"journal_title":null,"publication_types":["JournalArticle","Study"],"pubmed_pub_types":["Evaluation Study","Journal Article"],"s2_fields_of_study":["Medicine","Computer Science"],"reference_count":34,"citation_count":1549,"influential_citation_count":179,"is_open_access":true,"arxiv_categories":null,"arxiv_license":null,"arxiv_journal_ref":null,"mesh_headings":[{"d":"Algorithms","mj":true,"ui":"D000465"},{"d":"Artificial Intelligence","mj":true,"ui":"D001185"},{"d":"Image Interpretation, Computer-Assisted","mj":false,"qs":[{"q":"methods","mj":true,"ui":"Q000379"}],"ui":"D007090"},{"d":"Imaging, Three-Dimensional","mj":false,"qs":[{"q":"methods","mj":true,"ui":"Q000379"}],"ui":"D021621"},{"d":"Pattern Recognition, Automated","mj":false,"qs":[{"q":"methods","mj":true,"ui":"Q000379"}],"ui":"D010363"},{"d":"Photogrammetry","mj":false,"qs":[{"q":"methods","mj":true,"ui":"Q000379"}],"ui":"D010780"},{"d":"Subtraction Technique","mj":true,"ui":"D013382"}],"chemicals":null,"comments_corrections":null,"source_flags":5,"s2_open_access_pdf_url":"https://infoscience.epfl.ch/record/138785/files/tola_daisy_pami_1.pdf","s2_open_access_landing_url":"https://www.semanticscholar.org/paper/3ebc7ec2a9b89ebcc00328b11a1984d69a57f2a7","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":"In this paper, we introduce a local image descriptor, DAISY, which is very efficient to compute densely. We also present an EM-based algorithm to compute dense depth and occlusion maps from wide-baseline image pairs using this descriptor. This yields much better results in wide-baseline situations than the pixel and correlation-based algorithms that are commonly used in narrow-baseline stereo. Also, using a descriptor makes our algorithm robust against many photometric and geometric transformations. Our descriptor is inspired from earlier ones such as SIFT and GLOH but can be computed much faster for our purposes. Unlike SURF, which can also be computed efficiently at every pixel, it does not introduce artifacts that degrade the matching performance when used densely. It is important to note that our approach is the first algorithm that attempts to estimate dense depth maps from wide-baseline image pairs, and we show that it is a good one at that with many experiments for depth estimation accuracy, occlusion detection, and comparing it against other descriptors on laser-scanned ground truth scenes. We also tested our approach on a variety of indoor and outdoor scenes with different photometric and geometric transformations and our experiments support our claim to being robust against these.","claims":[{"public_id":"cl_848ac6493a6a70e2f2dfabf06152ae38","status":"active","text":"An EM-based algorithm using DAISY produces dense depth and occlusion maps from wide-baseline stereo pairs, achieving better results than pixel- and correlation-based algorithms used in narrow-baseline stereo.","confidence":0.95,"contributors":[{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["extraction"],"url":"https://sah.borca.ai/u/ezd9qvkvax"},{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["review"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":35,"public_id":"b2adb6bfad","public_label":"Anonymous (b2adb6bfad)","roles":["review"],"url":"https://sah.borca.ai/u/b2adb6bfad"},{"id":136,"public_id":"3c2apqe3ut","public_label":"Anonymous (3c2apqe3ut)","roles":["review"],"url":"https://sah.borca.ai/u/3c2apqe3ut"}],"url":"https://sah.borca.ai/claims/cl_848ac6493a6a70e2f2dfabf06152ae38"},{"public_id":"cl_a203d889861465fc70479cd131e4f39e","status":"active","text":"DAISY is a local image descriptor efficient for dense computation, inspired by SIFT and GLOH but computable much faster, and unlike SURF it does not introduce artifacts that degrade dense matching performance.","confidence":0.95,"contributors":[{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["extraction"],"url":"https://sah.borca.ai/u/ezd9qvkvax"},{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["review"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":35,"public_id":"b2adb6bfad","public_label":"Anonymous (b2adb6bfad)","roles":["review"],"url":"https://sah.borca.ai/u/b2adb6bfad"},{"id":136,"public_id":"3c2apqe3ut","public_label":"Anonymous (3c2apqe3ut)","roles":["review"],"url":"https://sah.borca.ai/u/3c2apqe3ut"}],"url":"https://sah.borca.ai/claims/cl_a203d889861465fc70479cd131e4f39e"},{"public_id":"cl_f44d02c56001543e1b4e6963c5f96297","status":"active","text":"The approach is robust against photometric and geometric transformations, as demonstrated on a variety of indoor and outdoor scenes.","confidence":0.9,"contributors":[{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["extraction"],"url":"https://sah.borca.ai/u/ezd9qvkvax"},{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["review"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":35,"public_id":"b2adb6bfad","public_label":"Anonymous (b2adb6bfad)","roles":["review"],"url":"https://sah.borca.ai/u/b2adb6bfad"},{"id":136,"public_id":"3c2apqe3ut","public_label":"Anonymous (3c2apqe3ut)","roles":["review"],"url":"https://sah.borca.ai/u/3c2apqe3ut"}],"url":"https://sah.borca.ai/claims/cl_f44d02c56001543e1b4e6963c5f96297"},{"public_id":"cl_21c945b333e3551deb1a25828dae13fe","status":"active","text":"This is the first algorithm that attempts to estimate dense depth maps from wide-baseline image pairs, with experiments showing good depth estimation accuracy and occlusion detection, and comparisons against other descriptors on laser-scanned ground truth scenes.","confidence":0.9,"contributors":[{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["extraction"],"url":"https://sah.borca.ai/u/ezd9qvkvax"},{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["review"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":35,"public_id":"b2adb6bfad","public_label":"Anonymous (b2adb6bfad)","roles":["review"],"url":"https://sah.borca.ai/u/b2adb6bfad"},{"id":136,"public_id":"3c2apqe3ut","public_label":"Anonymous (3c2apqe3ut)","roles":["review"],"url":"https://sah.borca.ai/u/3c2apqe3ut"}],"url":"https://sah.borca.ai/claims/cl_21c945b333e3551deb1a25828dae13fe"}],"concepts":[{"public_id":"co_0d1cfef5372c2fd7a6fed5b1a7fc7bb2","status":"active","name":"photometric transformations","description":"Image changes due to lighting, color, or camera response that the proposed approach is robust against.","types":["phenomenon"],"aliases":[],"contributors":[{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["extraction"],"url":"https://sah.borca.ai/u/ezd9qvkvax"},{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["review"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":35,"public_id":"b2adb6bfad","public_label":"Anonymous (b2adb6bfad)","roles":["review"],"url":"https://sah.borca.ai/u/b2adb6bfad"},{"id":136,"public_id":"3c2apqe3ut","public_label":"Anonymous (3c2apqe3ut)","roles":["review"],"url":"https://sah.borca.ai/u/3c2apqe3ut"}],"url":"https://sah.borca.ai/concepts/co_0d1cfef5372c2fd7a6fed5b1a7fc7bb2"},{"public_id":"co_0f433746c840d6a309b162cd8e6cbbac","status":"active","name":"occlusion maps","description":"Maps indicating occluded regions in stereo pairs, produced by the EM-based algorithm.","types":["output"],"aliases":["occlusion detection"],"contributors":[{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["extraction"],"url":"https://sah.borca.ai/u/ezd9qvkvax"},{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["review"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":35,"public_id":"b2adb6bfad","public_label":"Anonymous (b2adb6bfad)","roles":["review"],"url":"https://sah.borca.ai/u/b2adb6bfad"},{"id":136,"public_id":"3c2apqe3ut","public_label":"Anonymous (3c2apqe3ut)","roles":["review"],"url":"https://sah.borca.ai/u/3c2apqe3ut"}],"url":"https://sah.borca.ai/concepts/co_0f433746c840d6a309b162cd8e6cbbac"},{"public_id":"co_16351611a830b14ceb23b11fde4f3e7d","status":"active","name":"wide-baseline stereo","description":"Stereo matching with large baseline between views, where the proposed algorithm is applied.","types":["setting"],"aliases":["wide-baseline image pairs"],"contributors":[{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["extraction"],"url":"https://sah.borca.ai/u/ezd9qvkvax"},{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["review"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":35,"public_id":"b2adb6bfad","public_label":"Anonymous (b2adb6bfad)","roles":["review"],"url":"https://sah.borca.ai/u/b2adb6bfad"},{"id":136,"public_id":"3c2apqe3ut","public_label":"Anonymous (3c2apqe3ut)","roles":["review"],"url":"https://sah.borca.ai/u/3c2apqe3ut"}],"url":"https://sah.borca.ai/concepts/co_16351611a830b14ceb23b11fde4f3e7d"},{"public_id":"co_2821a91897bdd97cb8accf67d77328b6","status":"active","name":"narrow-baseline stereo","description":"Stereo matching with small baseline between views, where pixel- and correlation-based algorithms are common.","types":["setting"],"aliases":[],"contributors":[{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["extraction"],"url":"https://sah.borca.ai/u/ezd9qvkvax"},{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["review"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":35,"public_id":"b2adb6bfad","public_label":"Anonymous (b2adb6bfad)","roles":["review"],"url":"https://sah.borca.ai/u/b2adb6bfad"},{"id":136,"public_id":"3c2apqe3ut","public_label":"Anonymous (3c2apqe3ut)","roles":["review"],"url":"https://sah.borca.ai/u/3c2apqe3ut"}],"url":"https://sah.borca.ai/concepts/co_2821a91897bdd97cb8accf67d77328b6"},{"public_id":"co_4dc1412df65a2a480848d0b390e6dd36","status":"active","name":"SIFT","description":"A classic local image descriptor that inspired DAISY.","types":["method"],"aliases":["Scale-Invariant Feature Transform"],"contributors":[{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["extraction"],"url":"https://sah.borca.ai/u/ezd9qvkvax"},{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["review"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":35,"public_id":"b2adb6bfad","public_label":"Anonymous (b2adb6bfad)","roles":["review"],"url":"https://sah.borca.ai/u/b2adb6bfad"},{"id":136,"public_id":"3c2apqe3ut","public_label":"Anonymous (3c2apqe3ut)","roles":["review"],"url":"https://sah.borca.ai/u/3c2apqe3ut"}],"url":"https://sah.borca.ai/concepts/co_4dc1412df65a2a480848d0b390e6dd36"},{"public_id":"co_6a4143e1c14d8c71d092b29a9c4a2793","status":"active","name":"dense depth maps","description":"Per-pixel depth estimates produced by the proposed algorithm for wide-baseline stereo pairs.","types":["output"],"aliases":["dense depth estimation"],"contributors":[{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["extraction"],"url":"https://sah.borca.ai/u/ezd9qvkvax"},{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["review"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":35,"public_id":"b2adb6bfad","public_label":"Anonymous (b2adb6bfad)","roles":["review"],"url":"https://sah.borca.ai/u/b2adb6bfad"},{"id":136,"public_id":"3c2apqe3ut","public_label":"Anonymous (3c2apqe3ut)","roles":["review"],"url":"https://sah.borca.ai/u/3c2apqe3ut"}],"url":"https://sah.borca.ai/concepts/co_6a4143e1c14d8c71d092b29a9c4a2793"},{"public_id":"co_889be5ff4436bfe96c93bb4ef6247d3d","status":"active","name":"geometric transformations","description":"Image changes due to viewpoint, rotation, or scale that the proposed approach is robust against.","types":["phenomenon"],"aliases":[],"contributors":[{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["extraction"],"url":"https://sah.borca.ai/u/ezd9qvkvax"},{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["review"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":35,"public_id":"b2adb6bfad","public_label":"Anonymous (b2adb6bfad)","roles":["review"],"url":"https://sah.borca.ai/u/b2adb6bfad"},{"id":136,"public_id":"3c2apqe3ut","public_label":"Anonymous (3c2apqe3ut)","roles":["review"],"url":"https://sah.borca.ai/u/3c2apqe3ut"}],"url":"https://sah.borca.ai/concepts/co_889be5ff4436bfe96c93bb4ef6247d3d"},{"public_id":"co_939e8e3b7c07d5c1f5b624a21894a774","status":"active","name":"GLOH","description":"A local image descriptor that inspired DAISY.","types":["method"],"aliases":["Gradient Location and Orientation Histogram"],"contributors":[{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["extraction"],"url":"https://sah.borca.ai/u/ezd9qvkvax"},{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["review"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":35,"public_id":"b2adb6bfad","public_label":"Anonymous (b2adb6bfad)","roles":["review"],"url":"https://sah.borca.ai/u/b2adb6bfad"},{"id":136,"public_id":"3c2apqe3ut","public_label":"Anonymous (3c2apqe3ut)","roles":["review"],"url":"https://sah.borca.ai/u/3c2apqe3ut"}],"url":"https://sah.borca.ai/concepts/co_939e8e3b7c07d5c1f5b624a21894a774"},{"public_id":"co_a34ed8313247904512a82fd0d57ad6b5","status":"active","name":"EM-based algorithm","description":"An expectation-maximization algorithm that uses the DAISY descriptor to compute dense depth and occlusion maps from wide-baseline image pairs.","types":["method"],"aliases":["EM algorithm"],"contributors":[{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["extraction"],"url":"https://sah.borca.ai/u/ezd9qvkvax"},{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["review"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":35,"public_id":"b2adb6bfad","public_label":"Anonymous (b2adb6bfad)","roles":["review"],"url":"https://sah.borca.ai/u/b2adb6bfad"},{"id":136,"public_id":"3c2apqe3ut","public_label":"Anonymous (3c2apqe3ut)","roles":["review"],"url":"https://sah.borca.ai/u/3c2apqe3ut"}],"url":"https://sah.borca.ai/concepts/co_a34ed8313247904512a82fd0d57ad6b5"},{"public_id":"co_c1fe2d639931da661ce673ec4f82c3d7","status":"active","name":"laser-scanned ground truth scenes","description":"Reference scenes measured with laser scanning used to evaluate depth estimation accuracy and compare descriptors.","types":["dataset"],"aliases":["ground truth scenes"],"contributors":[{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["extraction"],"url":"https://sah.borca.ai/u/ezd9qvkvax"},{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["review"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":35,"public_id":"b2adb6bfad","public_label":"Anonymous (b2adb6bfad)","roles":["review"],"url":"https://sah.borca.ai/u/b2adb6bfad"},{"id":136,"public_id":"3c2apqe3ut","public_label":"Anonymous (3c2apqe3ut)","roles":["review"],"url":"https://sah.borca.ai/u/3c2apqe3ut"}],"url":"https://sah.borca.ai/concepts/co_c1fe2d639931da661ce673ec4f82c3d7"},{"public_id":"co_c3587fa73cad8754acf6a9a7985add04","status":"active","name":"SURF","description":"An efficient local image descriptor that can be computed at every pixel but introduces artifacts for dense matching, unlike DAISY.","types":["method"],"aliases":["Speeded Up Robust Features"],"contributors":[{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["extraction"],"url":"https://sah.borca.ai/u/ezd9qvkvax"},{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["review"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":35,"public_id":"b2adb6bfad","public_label":"Anonymous (b2adb6bfad)","roles":["review"],"url":"https://sah.borca.ai/u/b2adb6bfad"},{"id":136,"public_id":"3c2apqe3ut","public_label":"Anonymous (3c2apqe3ut)","roles":["review"],"url":"https://sah.borca.ai/u/3c2apqe3ut"}],"url":"https://sah.borca.ai/concepts/co_c3587fa73cad8754acf6a9a7985add04"},{"public_id":"co_d2320855ff0d9c86fbd47f16d60371f3","status":"active","name":"DAISY descriptor","description":"An efficient local image descriptor designed for dense computation, inspired by SIFT and GLOH but computable faster, avoiding artifacts seen with SURF.","types":["method"],"aliases":["DAISY"],"contributors":[{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["extraction"],"url":"https://sah.borca.ai/u/ezd9qvkvax"},{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["review"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":35,"public_id":"b2adb6bfad","public_label":"Anonymous (b2adb6bfad)","roles":["review"],"url":"https://sah.borca.ai/u/b2adb6bfad"},{"id":136,"public_id":"3c2apqe3ut","public_label":"Anonymous (3c2apqe3ut)","roles":["review"],"url":"https://sah.borca.ai/u/3c2apqe3ut"}],"url":"https://sah.borca.ai/concepts/co_d2320855ff0d9c86fbd47f16d60371f3"}],"external_ids":{"DOI":"10.1109/TPAMI.2009.77","ArXiv":null,"PubMed":20299707,"PubMedCentral":null,"MAG":2104853049,"DBLP":"journals/pami/TolaLF10","ACL":null},"open_access":{"is_open_access":true,"pdf_url":"https://infoscience.epfl.ch/record/138785/files/tola_daisy_pami_1.pdf","landing_url":"https://www.semanticscholar.org/paper/3ebc7ec2a9b89ebcc00328b11a1984d69a57f2a7","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":false,"full_text_source":null,"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":5},"paper_id":631260,"paper_uid":"a19ced85-f4d2-4c10-8192-29d351d646af","canonical_identity":{"paper_id":631260,"paper_uid":"a19ced85-f4d2-4c10-8192-29d351d646af","identity_status":"available","lookup_basis":"semantic_scholar_external_id","compatibility_path":"corpus_id"},"url":"https://sah.borca.ai/papers/206764601"}