| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 6.74 MB | Adobe PDF |
Authors
Advisor(s)
Abstract(s)
Esta tese de mestrado concentra-se no desenvolvimento de um programa avançado para a extração automática de eventos relavantes de transmissões de desportos motorizados. As corridas de Fórmula 1, com a sua audiência global e natureza dinâmica, servem como terreno ideal para esta investigação. O objetivo do projeto é melhorar a experiência de visualização para os entusiastas da Fórmula 1, ao mesmo tempo que fornece informações valiosas para fins de análise. A pesquisa começa com uma exploração do trabalho relacionado e depois aprofunda as bases teóricas do projeto, incluindo métodos de comparação de frames e técnicas de extração de padrões. A motivação do projeto centra-se na abordagem por métodos mais clássicos para a extração de eventos. A tese conclui resumindo as principais contribuições e discutindo as suas implicações, oferecendo uma compreensão abrangente do potencial impacto do projeto na extração de eventos em transmissões, particularmente em corridas de Fórmula 1.
This master’s thesis focuses on the development of an advanced program for the automatic extraction of relevant events from motor sports broadcasts. Formula 1 races, with their global audience and dynamic nature, serve as an ideal ground for this research. The project’s objective is to enhance the viewing experience for Formula 1 enthusiasts while providing valuable information for analysis purposes. The research begins with an exploration of related work and then delves into the theoretical foundations of the project, including frame comparison methods and pattern extraction techniques. The project’s motivation centers around the use of more classical methods for event extraction. The thesis concludes by summarizing the main contributions and discussing their implications, offering a comprehensive understanding of the potential impact of the project on event extraction in broadcasts, particularly in Formula 1 races.
This master’s thesis focuses on the development of an advanced program for the automatic extraction of relevant events from motor sports broadcasts. Formula 1 races, with their global audience and dynamic nature, serve as an ideal ground for this research. The project’s objective is to enhance the viewing experience for Formula 1 enthusiasts while providing valuable information for analysis purposes. The research begins with an exploration of related work and then delves into the theoretical foundations of the project, including frame comparison methods and pattern extraction techniques. The project’s motivation centers around the use of more classical methods for event extraction. The thesis concludes by summarizing the main contributions and discussing their implications, offering a comprehensive understanding of the potential impact of the project on event extraction in broadcasts, particularly in Formula 1 races.
Description
Projecto final para obtenção do Grau de Mestre em Engenharia Informática e Multimédia
Keywords
Processamento de vídeo Reconhecimento ótico de caracteres (OCR) Extração de padrões Computer vison Pattern recognition Optical Character Recognition (OCR)
Citation
MOTA, Catarina Maria Rodrigues - Deteção de eventos relevantes de automobilismo com base em Computer Vision. Lisboa: Instituto Superior de Engenharia de Lisboa, 2023. Dissertação de Mestrado.