| Revista: | Computación y Sistemas |
| Base de datos: | PERIÓDICA |
| Número de sistema: | 000457792 |
| ISSN: | 1405-5546 |
| Autores: | Díaz, Andrés1 Caicedo, Eduardo1 Paz, Lina2 Piniés, Pedro2 |
| Instituciones: | 1Universidad del Valle, Valle del Cauca. Colombia 2Intel Corporation, Santa Clara, California. Estados Unidos de América |
| Año: | 2020 |
| Periodo: | Ene-Mar |
| Volumen: | 24 |
| Número: | 1 |
| Paginación: | 221-239 |
| País: | México |
| Idioma: | Inglés |
| Tipo de documento: | Artículo |
| Enfoque: | Aplicado, descriptivo |
| Resumen en inglés | We present a system that improves the quality of noisy and incomplete depth maps captured with inexpensive range sensors. We use a model-based approach that measures the discrepancy between a model hypothesis and observed depth data. We represent the model hypothesis as a 3D level-set embedding function and the observed data as a point cloud coming from a segmented region associated to the object of interest. The discrepancy between the model and the observed data defines an objective function, that is minimized to obtain pose, scale and shape. The variation in shape of the object of interest is mapped with Gaussian Process Latent Variable Models GPLVM and the object pose is estimated using Lie algebra. The integration of a synthetic depth map, obtained from the optimal model, and the observed depth map is carried out with variational techniques. As a consequence we work in the observed space (depth space) rather than in a high dimensional volumetric space |
| Disciplinas: | Ciencias de la computación |
| Palabras clave: | Procesamiento de datos, Programación, Forma previa, Modelado 3D, Algoritmo de Levenberg-Marquardt, Algebra de Lie, Mapa de profundidad, Variaciones, Procesos gaussianos |
| Keyword: | Data processing, Programming, Shape prior, 3D modeling, Levenberg-Marquardt algorithm, Lie algebra, Depth map, Variations, Gaussian processes |
| Texto completo: | Texto completo (Ver HTML) Texto completo (Ver PDF) |
