"21230" . . "Recently it has been recognized that local image modeling in stereo matching brings more discriminable and stable matching features. We introduce a new disparity component matching approach, where the aggregation support regions are defined in disparity space, which guarantees invariance to input image view selection and independence on projective distortions. A rigorous ground-truth evaluation experiment showed that our approach is able to improve overall matching failure rate four-fold while the accuracy is preserved as compared to a standard fixed-size rectangular matching windows approach." . "Kostkov\u00E1, Jana" . . "2"^^ . "[8E05109E74BB]" . "S" . . "Praha" . "2"^^ . "RIV/68407700:21230/04:03103401!RIV08-MSM-21230___" . "Praha" . "Algoritmy p\u00E1rov\u00E1n\u00ED v po\u010D\u00EDta\u010Dov\u00E9m stereovid\u011Bn\u00ED"@cs . "2004-03-22+01:00"^^ . . . "364;365" . "\u0160\u00E1ra, Radim" . "\u010Cesk\u00E9 vysok\u00E9 u\u010Den\u00ED technick\u00E9 v Praze" . . "Proceedings of Workshop 2004" . "V ned\u00E1vn\u00E9 dob\u011B se uk\u00E1zalo, \u017Ee lok\u00E1ln\u00ED modelov\u00E1n\u00ED obrazu pro stereo vid\u011Bn\u00ED p\u0159in\u00E1\u0161\u00ED diskriminabiln\u011Bj\u0161\u00ED a stabiln\u011Bj\u0161\u00ED p\u00E1rovac\u00ED entity. My navrhujeme novou metodu p\u00E1rov\u00E1n\u00ED disparitn\u00EDch komponent, kde oblasti podpory jsou definov\u00E1ny v disparitn\u00EDm prostoru, co\u017E garantuje invarianci v\u016F\u010Di vstupn\u00EDm obraz\u016Fm a nez\u00E1vislost na projektivn\u00EDm zkreslen\u00ED. Rigor\u00F3zn\u00ED ground-truth experiment uk\u00E1zal, \u017Ee na\u0161e metoda je schopna zlep\u0161it celkovou chybovost p\u00E1rov\u00E1n\u00ED \u010Dty\u0159n\u00E1sobn\u011B a z\u00E1rove\u0148 zachovat jeho p\u0159esnost p\u0159i srovn\u00E1n\u00ED s klasick\u00FDm p\u00E1rov\u00E1n\u00EDm na z\u00E1klad\u011B pravo\u00FAhl\u00FDch, pevn\u011B dan\u00FDch oken."@cs . "Matching Algorithms in Computational Stereoscopic Vision" . . . . . "572413" . "Recently it has been recognized that local image modeling in stereo matching brings more discriminable and stable matching features. We introduce a new disparity component matching approach, where the aggregation support regions are defined in disparity space, which guarantees invariance to input image view selection and independence on projective distortions. A rigorous ground-truth evaluation experiment showed that our approach is able to improve overall matching failure rate four-fold while the accuracy is preserved as compared to a standard fixed-size rectangular matching windows approach."@en . . "2"^^ . . "computer vision; stereoscopic matching"@en . "Matching Algorithms in Computational Stereoscopic Vision"@en . "Algoritmy p\u00E1rov\u00E1n\u00ED v po\u010D\u00EDta\u010Dov\u00E9m stereovid\u011Bn\u00ED"@cs . . "Matching Algorithms in Computational Stereoscopic Vision"@en . . . "Matching Algorithms in Computational Stereoscopic Vision" . "80-01-02945-X" . "RIV/68407700:21230/04:03103401" .