Zpět na vyhledávání

Bc. Michal Červeňanský

Master's thesis

Automatic Image Annotation for Microstock Sites

Automatic Image Annotation for Microstock Sites
Abstract:
Táto práca sa zameriava na problém anotácie obrázkov vo fotobankovom priemysle. Na riešenie problému využíva najmodernejšie techniky strojového učenia pre detekciu objektov a klasifikáciu obrazu. Anotačný postup, optimalizovaný pre získanie anotácie v podobe kľúčových slov pre využitie vo fotobankách, je postavený na existujúcom nástroji MUFIN Annotation Framework. Pôvodnú presnosť anotácie MUFIN sa …more
Abstract:
This thesis addresses the issue of image annotation for the microstock industry. It attempts to bridge the gap between a real-life problem of image annotation and the state-of-the-art research of object detection and image classification techniques. Building upon the existing MUFIN Annotation Framework, we develop an annotation pipeline optimized for obtaining keyword annotation for microstock usage …more
 

Keywords

image annotation image classification object detection microstock convolutional neural networks MUFIN Annotation Framework seed keywords
 
Language used: English
Date on which the thesis was submitted / produced: 18. 5. 2021
Identifier: https://is.muni.cz/th/gisob/

Thesis defence

  • Date of defence: 25. 6. 2021
  • Supervisor: RNDr. Petra Budíková, Ph.D.
  • Reader: RNDr. Michal Batko, Ph.D.

Citation record

Cite this text

Full text of thesis

Contents of on-line thesis archive
Published in Theses:
  • světu
Other ways of accessing the text
Institution archiving the thesis and making it accessible: Masarykova univerzita, Fakulta informatiky
Složka Reference to the local database directory of the institution