Um método de interação com reconhecimento contínuo de gestos de toque para uso em smartwatch

• Ano de defesa: 2021
• Autor(a) principal: Nascimento, Thamer Horbylon
• Orientador(a): Soares, Fabrízzio Alphonsus Alves de Melo Nunes
• Banca de defesa: Soares, Fabrízzio Alphonsus Alves de Melo Nunes, Cardoso, Alexandre, Marques, Fátima de Lourdes dos Santos Nunes, Fernandes, Deborah Silva Alves, Bulcão Neto, Renato de Freitas
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Goiás
• Programa de Pós-Graduação: Programa de Pós-graduação em Ciência da Computação em Rede UFG/UFMS (INF)
• Departamento: Instituto de Informática - INF (RG)
• País: Brasil
• Palavras-chave em Português: Smartwatches; Relógios inteligentes; Métodos de interação; Reconhecimento contínuo de gestos; Interação com dispositivos
• Palavras-chave em Inglês: Smartwatches; Interaction methods; Continuous gesture recognition; Device interaction
• Área do conhecimento CNPq: CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO
• Link de acesso: http://repositorio.bc.ufg.br/tede/handle/tede/11844

Smartwatches are wearable and smart watch-shaped devices that aim to facilitate the daily lives of users, however, as they have small screens, the traditional mechanisms of interaction with touch-sensitive screens may not be efficient on these devices. Thus, it is important to develop new interaction methods that use the touchscreens of smartwatches. Therefore, in this work a method was developed that uses simple gestures, based on geometric shapes with continuous recognition of gestures on smartwatches touchscreens as an interaction mechanism. The results obtained with usability and/or experience tests of the method developed in case studies carried out in four segments are presented: text input in smartwatches and in virtual environments, interaction with platform games, video players and interactive movies and home appliances. During the development of this work, adaptations were made to the continuous gesture recognition algorithm: the recognition step was parallelized, which allows using threads to calculate the gesture probabilities; A technique was developed that allows detecting the change in gestures, allowing the user to initiate a new gesture without having to remove their finger from the screen. The results show that the continuous gesture recognition algorithm can be used in smartwatches and that with a small set of gestures it is possible to perform different actions. Thus, the developed method expands the possibilities of interaction with devices and environments using the smartwatches touch-sensitive screens and continuous recognition of gestures.