Tese de Dennis Giovani Balreira

Aluno: Dennis Giovani Balreira
Orientador: Prof. Dr. Marcelo Walter

Título: Automatic User-Specific Handwriting Synthesis
Linha de Pesquisa: Computação Gráfica e Visualização de Dados

Data: 17/05/2018
Horário: 14h30min.
Local: Prédio 43412 – Sala 215 (sala de videoconferência) do Instituto de Informática

Banca Examinadora:
– Prof. Dr. Cláudio Esperança (UFRJ – por videoconferência)
– Prof. Dr. Cláudio Rosito Jung (UFRGS)
– Prof. Dr. Eduardo Simões Lopes Gastal (UFRGS)

Presidente da Banca: Prof. Dr. Marcelo Walter

Abstract: Although nowadays people rely almost full time on digital text, the use of handwriting earned a special status for specific cases. We still learn to handwrite and use it as an identifying tool throughout our entire life. A particular field called handwriting synthesis generates renderings of text which resemble natural handwriting but are in fact synthesized by a model. This area has several uses in artistic applications, CAPTCHA generation and by providing new examples for handwriting recognition. It is still a challenging research area since it is difficult to mimic natural handwriting due to individual characteristics. Here we propose an online handwriting synthesis approach that uses public fonts as outputs. We intend to produce natural-looking results through a complete pipeline while also decreasing the computational cost when compared with state-of-the-art techniques. Based on our preliminary work, we intend to further investigate this approach. We are also exploring a new hardware which captures the users handwriting using common pen and paper, without introducing typical distortion produced by graphics tablets.

Keywords: Computer graphics, rendering, handwriting synthesis.