{"corpus_id":33230397,"paper_sha":"0010a842bd43073bd586776a9fae3452a796cae4","doi":"10.1007/978-3-319-75193-1_49","arxiv_id":"1708.07555","pmid":null,"pmcid":null,"mag_id":2749407104,"dblp_id":"conf/ciarp/NascimentoLBLN17","acl_id":null,"title":"A Robust Indoor Scene Recognition Method Based on Sparse Representation","year":2017,"publication_date":"2017-08-24","venue":"Iberoamerican Congress on Pattern Recognition","journal":{"name":null,"pages":"408-415","volume":null},"journal_issn":null,"journal_title":null,"publication_types":["JournalArticle"],"pubmed_pub_types":null,"s2_fields_of_study":["Computer Science"],"reference_count":21,"citation_count":17,"influential_citation_count":0,"is_open_access":false,"arxiv_categories":["cs.CV"],"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":null,"s2_open_access_landing_url":null,"s2_open_access_license":null,"s2_open_access_status":null,"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 present a robust method for scene recognition, which leverages Convolutional Neural Networks (CNNs) features and Sparse Coding setting by creating a new representation of indoor scenes. Although CNNs highly benefited the fields of computer vision and pattern recognition, convolutional layers adjust weights on a global-approach, which might lead to losing important local details such as objects and small structures. Our proposed scene representation relies on both: global features that mostly refers to environment’s structure, and local features that are sparsely combined to capture characteristics of common objects of a given scene. This new representation is based on fragments of the scene and leverages features extracted by CNNs. The experimental evaluation shows that the resulting representation outperforms previous scene recognition methods on Scene15 and MIT67 datasets, and performs competitively on SUN397, while being highly robust to perturbations in the input image such as noise and occlusion.","claims":[{"public_id":"cl_4fc7861c186919cca0b676da014fc1ee","status":"active","text":"A new indoor scene representation based on CNN features and sparse coding combines global structure cues with sparsely combined local object features.","confidence":0.98,"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_4fc7861c186919cca0b676da014fc1ee"},{"public_id":"cl_75bb9cf117ffdc21646337e00d118321","status":"active","text":"The resulting representation outperforms previous scene recognition methods on Scene15 and MIT67 datasets.","confidence":0.97,"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_75bb9cf117ffdc21646337e00d118321"},{"public_id":"cl_41b413a2b6d7c30a1efd5624c8872d3d","status":"active","text":"The resulting representation performs competitively on SUN397 and remains highly robust to input perturbations such as noise and occlusion.","confidence":0.96,"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_41b413a2b6d7c30a1efd5624c8872d3d"}],"concepts":[{"public_id":"co_1a9535d0fbd9f85dda5c543f1d229d30","status":"active","name":"sparse coding","description":"A representation strategy that encodes data as a sparse combination of basis elements or fragments.","types":["method"],"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_1a9535d0fbd9f85dda5c543f1d229d30"},{"public_id":"co_54bcd419c9d1df438570b49fe012f607","status":"active","name":"occlusion","description":"Partial blocking of image content used to test robustness of the scene representation.","types":["perturbation"],"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_54bcd419c9d1df438570b49fe012f607"},{"public_id":"co_5ec5ade9bf86b873b3263b96a06559ef","status":"active","name":"MIT67 dataset","description":"An indoor scene recognition benchmark containing 67 scene categories.","types":["dataset"],"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_5ec5ade9bf86b873b3263b96a06559ef"},{"public_id":"co_5f19fe59893a5d2444673ebc6038689c","status":"active","name":"Scene15 dataset","description":"An indoor scene recognition benchmark used for evaluation.","types":["dataset"],"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_5f19fe59893a5d2444673ebc6038689c"},{"public_id":"co_756021dc4e66cf459a67c4c035ceb492","status":"active","name":"common objects of a given scene","description":"Objects that are characteristic of a particular indoor scene category.","types":["visual concept"],"aliases":["objects of a given scene"],"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_756021dc4e66cf459a67c4c035ceb492"},{"public_id":"co_9989ee41098b11e233fbbfc3751d5ddf","status":"active","name":"new representation of indoor scenes","description":"A scene representation constructed from image fragments and CNN-extracted features for indoor scene recognition.","types":["representation"],"aliases":["scene representation","indoor scene representation"],"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_9989ee41098b11e233fbbfc3751d5ddf"},{"public_id":"co_9bf10a53a0cafa86cf3986f7c33a9453","status":"active","name":"previous scene recognition methods","description":"Earlier approaches used as baselines for comparison in indoor scene recognition.","types":["baseline"],"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_9bf10a53a0cafa86cf3986f7c33a9453"},{"public_id":"co_b18f8daaa4d3d2af57efc5095edaa22a","status":"active","name":"global features","description":"Features that capture the overall structure of a scene or environment.","types":["feature"],"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_b18f8daaa4d3d2af57efc5095edaa22a"},{"public_id":"co_b5368e6f89ee59ba24fac1af2b43a3b9","status":"active","name":"Convolutional Neural Networks","description":"Deep neural networks that use convolutional layers to extract visual features from images.","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_b5368e6f89ee59ba24fac1af2b43a3b9"},{"public_id":"co_efe2a4b8f6e614371ea86690bc7ea7c0","status":"active","name":"local features","description":"Features that capture local objects and small structures within a scene.","types":["feature"],"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_efe2a4b8f6e614371ea86690bc7ea7c0"},{"public_id":"co_f248427ac144c10e5c038e9f872d769b","status":"active","name":"SUN397","description":"A large-scale scene recognition benchmark used for evaluation.","types":["dataset"],"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_f248427ac144c10e5c038e9f872d769b"},{"public_id":"co_f55935c3287466cab090aaca920a377d","status":"active","name":"noise","description":"Random image corruption used to test robustness of the scene representation.","types":["perturbation"],"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_f55935c3287466cab090aaca920a377d"}],"external_ids":{"DOI":"10.1007/978-3-319-75193-1_49","ArXiv":"1708.07555","PubMed":null,"PubMedCentral":null,"MAG":2749407104,"DBLP":"conf/ciarp/NascimentoLBLN17","ACL":null},"open_access":{"is_open_access":true,"pdf_url":"https://arxiv.org/pdf/1708.07555","landing_url":"https://arxiv.org/abs/1708.07555","source":"arxiv","pdf_url_source":"derived_arxiv","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","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":631111,"paper_uid":"5b0ff37d-7657-496e-9d7d-5d91fb27a93e","canonical_identity":{"paper_id":631111,"paper_uid":"5b0ff37d-7657-496e-9d7d-5d91fb27a93e","identity_status":"available","lookup_basis":"semantic_scholar_external_id","compatibility_path":"corpus_id"},"url":"https://sah.borca.ai/papers/33230397"}