magistrsko delo
Povzetek
Ljudje pogosto delamo napake, zato vedno bolj stremimo k avtomatizaciji vseh aspektov življena. Tudi področje športnih panog ni izjema. Več kot desetletje nazaj so tekme analizirali ljudje, sedaj za ekipe to počne umetna inteligenca. Zaradi njenega hitrega razvoja v zadnjem desetletju, so nevronske mreže vedno bolj natančne, hitre in v metrikah na določenih področjih dosegajo boljše rezultate kot ljudje. Motivacija za izdelavo magistrske naloge je razvoj algoritma, ki bi zaznaval statistiko med prenosom lige NBA. Prav tako bi vizualno z obogateno resničnostjo uporabniku pomagal pri razumevanju igre. Za cilj smo si zadali izdelavo algoritma, ki s čim višjo natančnostjo sledi igralcem na igrišču in detektira njihove akcije. V okviru magistrske naloge bomo tako raziskali sodobne in učinkovite metode znotraj problemske domene ter jih prilagodili za naš primer. Razvili bomo algoritem, ki uspešno zaznava igralce na igrišču in jih klasificirali v pripadajočo ekipo z uporabo nevronskih mrež. Z uporabo homografske transformacije bomo prenesli njihovo lokacijo v dvodimenzionalni prostor na igrišču ter definirali nov algoritem za zaznavo akcij, ki jih izvajajo igralci in pridobili potrebne podatke za beleženje njihovih statistik. Tekom implementacije algoritma smo preizkusili različne metode reševanja problema. Ugotovili smo njihovo uspešnost in analizirali njihove prednosti ter slabosti.
Ključne besede
homografija;detekcija;avtomatsko beleženje statistike;analiza video posnetkov;nevronske mreže;klasifikacija;sledenje objektov;magisteriji;
Podatki
Jezik: |
Slovenski jezik |
Leto izida: |
2022 |
Tipologija: |
2.09 - Magistrsko delo |
Organizacija: |
UL FRI - Fakulteta za računalništvo in informatiko |
Založnik: |
[A. Veršnik] |
UDK: |
004.93:004.8:796.323.2(043.2) |
COBISS: |
150891267
|
Št. ogledov: |
97 |
Št. prenosov: |
31 |
Ocena: |
0 (0 glasov) |
Metapodatki: |
|
Ostali podatki
Sekundarni jezik: |
Angleški jezik |
Sekundarni naslov: |
Automatic tracking of NBA statistics from a live broadcast |
Sekundarni povzetek: |
People often make mistakes, that is why we try to automate every aspect of our lives. The field of sports is not an exception. While only a bit more than a decade ago people analyzed games, this is now done by artificial intelligence. Because of its fast development in the last ten years, neural networks are now faster, more accurate and in those metrics better then its human counterparts in some fields. The motivation for the master's thesis is thus to develop an algorithm, that can detect player statistics during an NBA broadcast. It would also help the user to better understand the game with the use of augmented reality. The aim was to create an algorithm that could detect the players on the court and track their actions with the highest accuracy possible. In the framework of the master's thesis we studied modern and effective methods in the knowledge domain. We developed an algorithm that could successfully detect players on the court, and classify them to their respective team with a neural network. With the use of a the homography transformation we moved the positions of the players to a two dimensional space on the court. We defined a new algorithm to detect the actions of the players, and thus their statistics. During the implementation of the algorithm we tried different methods to solve the problem. We analyzed their effectiveness and discussed their strengths and weaknesses. |
Sekundarne ključne besede: |
homography;computer vision;detection;automatic tracking of statistics;basketball;video analysis;neural networks;classification;object tracking;computer science;computer and information science;master's degree;Računalniški vid;Umetna inteligenca;Košarka;Računalništvo;Univerzitetna in visokošolska dela; |
Vrsta dela (COBISS): |
Magistrsko delo/naloga |
Študijski program: |
1000471 |
Konec prepovedi (OpenAIRE): |
1970-01-01 |
Komentar na gradivo: |
Univ. v Ljubljani, Fak. za računalništvo in informatiko |
Strani: |
62 str. |
ID: |
18367743 |