Revista: | Computación y sistemas |
Base de datos: | |
Número de sistema: | 000560475 |
ISSN: | 1405-5546 |
Autores: | Valdez Rodríguez, José E1 Calvo, Hiram1 Felipe Riverón, Edgardo M1 |
Instituciones: | 1Instituto Politécnico Nacional, Centro de Investigación en Computación, Ciudad de México. México |
Año: | 2020 |
Periodo: | Abr-Jun |
Volumen: | 24 |
Número: | 2 |
Paginación: | 439-451 |
País: | México |
Idioma: | Inglés |
Tipo de documento: | Artículo |
Resumen en inglés | Depth reconstruction from single monocular images has been a challenging task due to the complexity and the quantity of depth cues that images have. Convolutional Neural Networks (CNN) have been successfully used to reconstruct depth of general object scenes; however, proposed works use several stages of training which make this process more complex and time consuming. As we aim to build a computational efficient model, we focus on single-stage training CNN. In this paper, we propose five different models for solving this task, ranging from a simple convolutional network, to one with residual, convolutional, refinement and upsampling layers. We compare our models with the current state of the art in depth reconstruction and measure depth reconstruction error for different datasets (KITTI, NYU), obtaining improvements in both global and local error measures. |
Disciplinas: | Ciencias de la computación |
Palabras clave: | Inteligencia artificial |
Keyword: | Depth reconstruction, Convolutional neural networks, Single stage training, Embedded refinement layer, Stereo matching, Artificial intelligence |
Texto completo: | Texto completo (Ver HTML) Texto completo (Ver PDF) |