Image Generation from Sketches and Text-Guided Attribute Edition

Sumiri Fernandez D.M.; Ochoa-Luna J. · Communications in Computer and Information Science (2023) · DOI: 10.1007/978-3-031-35445-8_7

Resumen

Every day new Artificial Intelligence models are created, and existing ones are modified or combined to extend their range of applications and tasks they can solve. This paper presents a novel approach that combines Natural Language Processing and generative networks to generate images from sketches and descriptions in natural language. We present a pipeline that was followed to recondition the generative network. It includes the processed text that will give the context to the sketch used for the generation. Finally, the model and the generated images are evaluated and compared using benchmark data sets, and promising results are reported.

Instituciones: Universidad Católica San Pablo