Recognising people by means of clothing or other non-invasive factors

Identifier:  TFM:2106

Handle: https://hdl.handle.net/20.500.11797/TFM2106

Authors:  Koeckerbauer, Daniel-Markus

Abstract:
This thesis titled explores the development of a non-invasive person recognition system. This system leverages attributes such as clothing and height to enhance ski access control mechanisms, specifically for SKIDATA, the project's sponsor. The thesis covers the theoretical foundation of neural networks, particularly Siamese networks, and details the creation and processing of datasets to train the AI model. It also covers challenges of the process. The approach aims to improve the efficiency and accuracy of recognizing individuals at ski resorts, addressing the legal challenges posed by traditional biometric methods.