{"id":333992,"date":"2016-12-08T17:49:50","date_gmt":"2016-12-09T01:49:50","guid":{"rendered":"https:\/\/www.microsoft.com\/en-us\/research\/?post_type=msr-research-item&#038;p=333992"},"modified":"2018-10-16T20:12:34","modified_gmt":"2018-10-17T03:12:34","slug":"segmenting-textured-3d-surfaces-using-spacefrequency-representation","status":"publish","type":"msr-research-item","link":"https:\/\/www.microsoft.com\/en-us\/research\/publication\/segmenting-textured-3d-surfaces-using-spacefrequency-representation\/","title":{"rendered":"Segmenting Textured 3D Surfaces Using the Space\/Frequency Representation"},"content":{"rendered":"<p>Segmenting <b>3D <\/b>textured surfaces is critical for general image understanding. Unfortunately, current efforts at automatically understanding image texture are based on assumptions that make this goal impossible. Texture segmentation research assumes that the textures are flat and viewed from the front, while shape-from-texture work assumes that the textures have already been segmented. This deadlock means that none of these algorithms will work reliably on images <b>of 3D <\/b>textured surfaces.<\/p>\n<p>We have developed an algorithm that can segment an image containing nonfrontally viewed, planar, periodic textures. We use the spectrogram (local power spectrum) to compute <b>local <\/b>surface normals from small regions of the image. This algorithm does not require unreliable image feature detection. Based on these surface normals, we compute a \u201cfrontalized\u201d version of the local power spectrum which shows what the region\u2019s power spectrum would look like if viewed from the front. If neighboring regions have similar frontalized power spectra, they are merged. The merge criteria is based on a description length formula. We demonstrate the segmentation on images with real textures. To our knowledge, this is the first program that can segment <b>3D <\/b>textured surfaces by explicitly accounting for shape effects.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Segmenting 3D textured surfaces is critical for general image understanding. Unfortunately, current efforts at automatically understanding image texture are based on assumptions that make this goal impossible. Texture segmentation research assumes that the textures are flat and viewed from the front, while shape-from-texture work assumes that the textures have already been segmented. This deadlock means [&hellip;]<\/p>\n","protected":false},"featured_media":0,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"_classifai_error":"","msr-author-ordering":null,"msr_publishername":"","msr_publisher_other":"","msr_booktitle":"","msr_chapter":"","msr_edition":"Spatial Vision 1994, special issue on Texture Perception and Attention","msr_editors":"","msr_how_published":"","msr_isbn":"","msr_issue":"2","msr_journal":"","msr_number":"","msr_organization":"","msr_pages_string":"281-308","msr_page_range_start":"281","msr_page_range_end":"308","msr_series":"","msr_volume":"8","msr_copyright":"This research was sponsored by the Avionics Laboratory, Wright Research and Development Center, Aeronautical Systems Division (AFSC), U. S. Air Force, Wright-Patterson AFB, OH 45433-6543 under Contract F33615-90-C-1465, Arpa Order No. 7597. This first author was supported by NASA under the Graduate Student Researchers Program, Grant No. NGT-50423. The views and conclusions contained in this document are those of the authors and should not be interpreted as representing the official policies, either expressed or implied, of the U.S. Government.","msr_conference_name":"","msr_doi":"","msr_arxiv_id":"","msr_s2_paper_id":"","msr_mag_id":"","msr_pubmed_id":"","msr_other_authors":"","msr_other_contributors":"","msr_speaker":"","msr_award":"","msr_affiliation":"","msr_institution":"","msr_host":"","msr_version":"","msr_duration":"","msr_original_fields_of_study":"","msr_release_tracker_id":"","msr_s2_match_type":"","msr_citation_count_updated":"","msr_published_date":"1994-01-01","msr_highlight_text":"","msr_notes":"Spatial Vision 1994, special issue on Texture Perception and Attention","msr_longbiography":"","msr_publicationurl":"","msr_external_url":"","msr_secondary_video_url":"","msr_conference_url":"","msr_journal_url":"","msr_s2_pdf_url":"","msr_year":0,"msr_citation_count":0,"msr_influential_citations":0,"msr_reference_count":0,"msr_s2_match_confidence":0,"msr_microsoftintellectualproperty":true,"msr_s2_open_access":false,"msr_s2_author_ids":[],"msr_pub_ids":[],"msr_hide_image_in_river":0,"footnotes":""},"msr-research-highlight":[],"research-area":[13563],"msr-publication-type":[193724],"msr-publisher":[],"msr-focus-area":[],"msr-locale":[268875],"msr-post-option":[],"msr-field-of-study":[],"msr-conference":[],"msr-journal":[],"msr-impact-theme":[],"msr-pillar":[],"class_list":["post-333992","msr-research-item","type-msr-research-item","status-publish","hentry","msr-research-area-data-platform-analytics","msr-locale-en_us"],"msr_publishername":"","msr_edition":"Spatial Vision 1994, special issue on Texture Perception and Attention","msr_affiliation":"","msr_published_date":"1994-01-01","msr_host":"","msr_duration":"","msr_version":"","msr_speaker":"","msr_other_contributors":"","msr_booktitle":"","msr_pages_string":"281-308","msr_chapter":"","msr_isbn":"","msr_journal":"","msr_volume":"8","msr_number":"","msr_editors":"","msr_series":"","msr_issue":"2","msr_organization":"","msr_how_published":"","msr_notes":"Spatial Vision 1994, special issue on Texture Perception and Attention","msr_highlight_text":"","msr_release_tracker_id":"","msr_original_fields_of_study":"","msr_download_urls":"","msr_external_url":"","msr_secondary_video_url":"","msr_longbiography":"","msr_microsoftintellectualproperty":1,"msr_main_download":"333995","msr_publicationurl":"","msr_doi":"","msr_publication_uploader":[{"type":"file","title":"Segmenting Textured 3D Surfaces Using the Space\/Frequency Representation","viewUrl":"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2016\/12\/krumm_john_1994_2.pdf","id":333995,"label_id":0}],"msr_related_uploader":"","msr_citation_count":0,"msr_citation_count_updated":"","msr_s2_paper_id":"","msr_influential_citations":0,"msr_reference_count":0,"msr_arxiv_id":"","msr_s2_author_ids":[],"msr_s2_open_access":false,"msr_s2_pdf_url":null,"msr_attachments":[],"msr-author-ordering":[{"type":"user_nicename","value":"jckrumm","user_id":32203,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=jckrumm"},{"type":"text","value":"Steven A. Shafer","user_id":0,"rest_url":false}],"msr_impact_theme":[],"msr_research_lab":[],"msr_event":[],"msr_group":[],"msr_project":[],"publication":[],"video":[],"msr-tool":[],"msr_publication_type":"miscellaneous","related_content":[],"_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/333992","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item"}],"about":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/types\/msr-research-item"}],"version-history":[{"count":1,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/333992\/revisions"}],"predecessor-version":[{"id":524651,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/333992\/revisions\/524651"}],"wp:attachment":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media?parent=333992"}],"wp:term":[{"taxonomy":"msr-research-highlight","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-highlight?post=333992"},{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=333992"},{"taxonomy":"msr-publication-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-publication-type?post=333992"},{"taxonomy":"msr-publisher","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-publisher?post=333992"},{"taxonomy":"msr-focus-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-focus-area?post=333992"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=333992"},{"taxonomy":"msr-post-option","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-post-option?post=333992"},{"taxonomy":"msr-field-of-study","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-field-of-study?post=333992"},{"taxonomy":"msr-conference","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-conference?post=333992"},{"taxonomy":"msr-journal","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-journal?post=333992"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=333992"},{"taxonomy":"msr-pillar","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-pillar?post=333992"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}