engleski

Development of a Visual Search Engine

Visual search has the potential to change the way we interact with the world around us because it uses real-world images as the stimuli for online searches. Modern visual search technology uses a combination of computer vision and machine learning to understand the content and context of these images and return a list of related results. In this thesis, the state-of-the-art methods for developing a visual search engine are elaborated and discussed. The aim is to design and implement an example of visual search engine using machine learning algorithms.

visual search engine ; content based image retrieval ; machine learning ; deep neural networks ; transfer learning

nije evidentirano