{"id":321,"date":"2024-09-20T04:11:43","date_gmt":"2024-09-20T02:11:43","guid":{"rendered":"https:\/\/ics.science.upjs.sk\/ano\/?page_id=321"},"modified":"2026-10-22T09:28:51","modified_gmt":"2026-10-22T07:28:51","slug":"prednasky","status":"publish","type":"page","link":"https:\/\/ics.science.upjs.sk\/ano\/prednasky\/","title":{"rendered":"Predn\u00e1\u0161ky"},"content":{"rendered":"<section id=\"harmonogram\" class=\"level2\">\n<ul>\n<li>17.09: <strong>01 \u00davod &#8211; <\/strong>\u00davod do po\u010d\u00edta\u010dov\u00e9ho videnia.<\/li>\n<li>24.09: <strong>02 Bin\u00e1rny obraz a morfol\u00f3gia &#8211; <\/strong>Farebn\u00fd, \u0161edot\u00f3nov\u00fd, bin\u00e1rny obraz. Prahovanie, histogram, vyhladzovanie histogramu. Matematick\u00e1 morfol\u00f3gia.<\/li>\n<li>01.10: <strong>03 \u0160um a konvol\u00facia <\/strong>\u0160um, odstra\u0148ovanie \u0161umu,diffusion model (generovanie obr\u00e1zkov). Filtrovanie, konvol\u00facia,\u00a0<em>indoor positioning &#8211; grid a konvol\u00facia.<\/em><\/li>\n<li>08.10: bez predn\u00e1\u0161ky<\/li>\n<li>15.10: <strong>04 Transform\u00e1cie &#8211; <\/strong>Filtrovanie vo frekven\u010dnej oblasti, Fourierov\u00e1 tranform\u00e1cia, konvolu\u010dn\u00e1 veta, s\u00ednusoida, vzorkovanie, aliasing. Met\u00f3da najmen\u0161\u00edch \u0161tvorcov, RANSAC. Hough transform\u00e1cia na n\u00e1jdenie \u010diar a kruhov.<\/li>\n<li>22.10: <strong>05 Detekcia hr\u00e1n &#8211; <\/strong>Detekcia hr\u00e1n, gradient, laplacian, Cannyho hranov\u00fd detektor, detekcia rohov.<\/li>\n<li>29.10: <strong>06 Segment\u00e1cia a rozpozn\u00e1vanie &#8211; <\/strong>Segment\u00e1cia obrazu a rozpozn\u00e1vanie. Klastrovanie (k\u00ad\u2011means, meanshift). Grabcut. Detekcia tv\u00e1re, Haar features.<\/li>\n<li>05.11: <strong>07 Pr\u00edznaky &#8211; <\/strong>Pr\u00edznaky. Blob detekcia. SIFT detektor a deskriptor. Bag of visual words. Geometrick\u00e9 transform\u00e1cie<strong>.<\/strong><\/li>\n<li>12.11: bez predn\u00e1\u0161ky<\/li>\n<li>19.11: bez predn\u00e1\u0161ky\u00a0 (rezerva ak nie\u010do predt\u00fdm odpadlo)<\/li>\n<li>26.11: <strong>08 Modern\u00e9 CV &#8211; <\/strong>Multimodal input, Visual embedding, image retrieval, modern\u00e9 detektory &#8211; yolo, active learning, self-supervised learning (dino), SAM, CLIP, pose estimation<\/li>\n<li>03.12: <strong>09 Tracking &#8211;\u00a0<\/strong>Sledovanie objektu na sekvenci\u00ed obr\u00e1zkov, mixture of gaussians, template matching, tracking &#8211;\u00a0SOT vs.\u00a0MOT, tracking-by-detection, SORT, Kalman filter.<\/li>\n<li>10.12: <strong>10 3D &#8211;\u00a0<\/strong>Vznik obrazu \u2013 dierkov\u00e1 kamera. Projekcia z 3D do 2D, vonkaj\u0161ia a vn\u00fatorn\u00e1 matica, kalibr\u00e1cia kamery, epipol\u00e1rna geometria, h\u013abka obrazu. Structure from Motion.<\/li>\n<li>17.12: bez predn\u00e1\u0161ky\u00a0 (mo\u017eno predterm\u00edn)<\/li>\n<\/ul>\n<\/section>\n<section id=\"osnova-predmetu\" class=\"level2\"><\/section>\n<hr \/>\n<h3>1. \u00davod do po\u010d\u00edta\u010dov\u00e9ho videnia<\/h3>\n<ul>\n<li><a href=\"http:\/\/karpathy.github.io\/2012\/10\/22\/state-of-computer-vision\/\">Obama na v\u00e1he<\/a>\u00a0\u2013 blog \u010dl\u00e1nok o stave AI pri poh\u013eade na fotku<\/li>\n<li><a href=\"https:\/\/cloud.google.com\/vision\">Google Cloud Vision<\/a>\u00a0\u2013 d\u00e1 sa vysk\u00fa\u0161a\u0165 rozpozn\u00e1vanie obr\u00e1zkov<\/li>\n<\/ul>\n<h3>2. Farebn\u00fd, \u0161edot\u00f3nov\u00fd, bin\u00e1rny obraz. Prahovanie, histogram, vyhladzovanie histogramu. Matematick\u00e1 morfol\u00f3gia.<\/h3>\n<p><em>Tento obsah je pokryt\u00fd v skript\u00e1ch (okrem vyhladzovania histogramu &#8211; to bolo na cvi\u010deniach aj s pr\u00edkladmi)<\/em><em>.<\/em><\/p>\n<ul>\n<li><a href=\"https:\/\/ics.science.upjs.sk\/ano\/wp-content\/uploads\/sites\/17\/2025\/09\/02-binarne-obrazy.pptx.pdf\">slajdy<\/a><\/li>\n<li><a href=\"https:\/\/www.redblobgames.com\/grids\/hexagons\/\">\u0160es\u0165uholn\u00edky<\/a><\/li>\n<li><a href=\"https:\/\/playgameoflife.com\/\">Game of life<\/a><\/li>\n<li><a href=\"https:\/\/docs.opencv.org\/4.13.0\/d5\/daf\/tutorial_py_histogram_equalization.html\">vyhladzovanie histogramu<\/a><\/li>\n<li><a href=\"https:\/\/ics.science.upjs.sk\/ano\/wp-content\/uploads\/sites\/17\/2025\/09\/nFtqv8e.jpg\">obr\u00e1zok<\/a> + <a href=\"https:\/\/www.gimp.org\/\">GIMP<\/a><\/li>\n<\/ul>\n<h3>3. \u0160um, odstra\u0148ovanie \u0161umu, diffusion model (generovanie obr\u00e1zkov). Filtrovanie, konvol\u00facia, indoor positioning &#8211; grid a konvol\u00facia.<\/h3>\n<p><em>Tento obsah nie je v skript\u00e1ch. Slajdy s\u00fa len pomocn\u00e9 k predn\u00e1\u0161ke. Indoor lokaliz\u00e1cia je bonusov\u00fd obsah. Pre \u0161um a konvol\u00faciu odpor\u00fa\u010dam pozrie\u0165 na zoznam literat\u00fary.\u00a0<\/em><\/p>\n<ul>\n<li><a href=\"https:\/\/ics.science.upjs.sk\/ano\/wp-content\/uploads\/sites\/17\/2022\/10\/03-filtrovanie.pdf\">slajdy<\/a><\/li>\n<li><a href=\"https:\/\/www.cambridgeincolour.com\/tutorials\/gamma-correction.htm#:~:text=Gamma%20correction%20is%20sometimes%20specified,1%2F1.5%20%3D%201.0).\">Gamma korekcia<\/a><\/li>\n<li><a href=\"https:\/\/en.wikipedia.org\/wiki\/Convolution\">konvol\u00facia<\/a>\u00a0\u2013 vizualiz\u00e1cia na wikip\u00e9dii<\/li>\n<li><a href=\"https:\/\/ics.upjs.sk\/~opiela\/iot\/uvod-do-iot\/\">Indoor lokaliz\u00e1cia<\/a> (2021) + <a href=\"https:\/\/ics.science.upjs.sk\/indora\/\">aktu\u00e1lny v\u00fdskum<\/a><\/li>\n<li>Diffusion model &#8211; <a href=\"https:\/\/upload.wikimedia.org\/wikipedia\/commons\/f\/f6\/Stable_Diffusion_architecture.png?utm_source=sk.wikipedia.org&amp;utm_campaign=index&amp;utm_content=original\">architekt\u00fara<\/a>, <a href=\"https:\/\/sk.wikipedia.org\/wiki\/S%C3%BAbor:X-Y_plot_of_algorithmically-generated_AI_art_of_European-style_castle_in_Japan_demonstrating_DDIM_diffusion_steps.png\">proces<\/a>, <a href=\"https:\/\/stablediffusionweb.com\/\">web<\/a>, <a href=\"https:\/\/github.com\/compvis\/stable-diffusion\">github<\/a>, <a href=\"https:\/\/ommer-lab.com\/research\/latent-diffusion-models\/\">\u010dl\u00e1nok<\/a><\/li>\n<\/ul>\n<h3>4. Filtrovanie vo frekven\u010dnej oblasti. Hough transform\u00e1cia.<\/h3>\n<p><em>Obsah nem\u00e1m pokryt\u00fd v skript\u00e1ch. Ide o viacero osobitn\u00fdch t\u00e9m.\u00a0<\/em><\/p>\n<ul>\n<li><a href=\"https:\/\/www.geogebra.org\/m\/wUanseCs\">Geogebra<\/a> &#8211; square wave dekompoz\u00edcia pomocou fourierovej transform\u00e1cie<\/li>\n<li><a href=\"https:\/\/en.wikipedia.org\/wiki\/File:Fourier_transform_time_and_frequency_domains.gif\">Fourierov\u00e1 transform\u00e1cia<\/a> &#8211; anim\u00e1cia + <a href=\"https:\/\/upload.wikimedia.org\/wikipedia\/commons\/6\/6b\/SquareWaveFourierArrows.gif\">anim\u00e1cia<\/a> pomocou \u0161\u00edpok<\/li>\n<li><a href=\"https:\/\/www.youtube.com\/watch?v=ebfi7qOFLuo\">Hough transform\u00e1cia<\/a> &#8211; youtube video, demo<\/li>\n<\/ul>\n<h3>5. Detekcia hr\u00e1n, gradient, laplacian, Cannyho hranov\u00fd detektor, detekcia rohov.<\/h3>\n<p><em>\u00davod o hran\u00e1ch a gradient je pokryt\u00fd v skript\u00e1ch. Materi\u00e1ly k tejto t\u00e9me s\u00fa najlep\u0161ie vo\u00a0 First Principle\u00a0(viac v zozname literat\u00fary).<\/em><\/p>\n<ul>\n<li><a href=\"https:\/\/opencv24-python-tutorials.readthedocs.io\/en\/latest\/py_tutorials\/py_imgproc\/py_canny\/py_canny.html\">Canny edge detection<\/a>\u00a0\u2013 opencv<\/li>\n<li><a href=\"https:\/\/cave.cs.columbia.edu\/Statics\/monographs\/Edge%20Detection%20FPCV-2-1.pdf\">First Principle &#8211; Edge detection<\/a> &#8211; lecture notes<\/li>\n<li><a href=\"https:\/\/ics.science.upjs.sk\/ano\/wp-content\/uploads\/sites\/17\/2026\/10\/edge.pdf\">Edge detection<\/a> &#8211; prezent\u00e1cia (extrahovan\u00e1 z lecture notes)<\/li>\n<\/ul>\n<h3><\/h3>\n","protected":false},"excerpt":{"rendered":"<p>17.09: 01 \u00davod &#8211; \u00davod do po\u010d\u00edta\u010dov\u00e9ho videnia. 24.09: 02 Bin\u00e1rny obraz a morfol\u00f3gia &#8211; Farebn\u00fd, \u0161edot\u00f3nov\u00fd, bin\u00e1rny obraz. Prahovanie, histogram, vyhladzovanie histogramu. Matematick\u00e1 morfol\u00f3gia. 01.10: 03 \u0160um a konvol\u00facia&hellip;<\/p>\n","protected":false},"author":22,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"footnotes":""},"class_list":["post-321","page","type-page","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/ics.science.upjs.sk\/ano\/wp-json\/wp\/v2\/pages\/321","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/ics.science.upjs.sk\/ano\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/ics.science.upjs.sk\/ano\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/ics.science.upjs.sk\/ano\/wp-json\/wp\/v2\/users\/22"}],"replies":[{"embeddable":true,"href":"https:\/\/ics.science.upjs.sk\/ano\/wp-json\/wp\/v2\/comments?post=321"}],"version-history":[{"count":49,"href":"https:\/\/ics.science.upjs.sk\/ano\/wp-json\/wp\/v2\/pages\/321\/revisions"}],"predecessor-version":[{"id":577,"href":"https:\/\/ics.science.upjs.sk\/ano\/wp-json\/wp\/v2\/pages\/321\/revisions\/577"}],"wp:attachment":[{"href":"https:\/\/ics.science.upjs.sk\/ano\/wp-json\/wp\/v2\/media?parent=321"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}