Parallel Implementation of a Robust Optical Flow Technique

Autores que no pertenecientes al CTIM

Sánchez Pérez, Javier; Monzón López, Nelson; Salgado de la Nuez, Agustín J

---

Línea de investigación:

Flujo óptico

---

Año:

2012

---

Tipo de publicación:

Reporte técnico

---

Institución:

Universidad de Las Palmas de Gran Canaria

---

Palabras claves:

Optical flow, Variational methods, PDE, Multi-core, OpenMP

---

Número:

1

---

ISSN:

2254-2353

---

Mes:

Marzo

---

BibTex:

@techreport{CTIM_1, author = "Javier S{\'a}nchez P{\'e}rez and Nelson Monz{\'o}n L{\'o}pez and Agust{\'i}n J. Salgado de la Nuez", abstract = "The accuracy and performance of current variational optical flow methods have considerably increased during the last years. The complexity of these techniques is high and enough care has to be taken for the implementation. The aim of this work is to present a comprehensible implementation of recent variational optical flow methods. We start with an energy model that relies on brightness and gradient constancy terms and a flow-based smoothness term. We minimize this energy model and derive an efficient implicit numerical scheme. In the experimental results, we evaluate the accuracy and performance of this implementation with the Middlebury benchmark database. We show that it is a competitive solution with respect to current methods in the literature. In order to increase the performance, we use a simple strategy to parallelize the execution on multi-core processors", booktitle = "CTIM Technical Report", institution = "Universidad de Las Palmas de Gran Canaria", issn = "2254-2353", journal = "CTIM Technical Report", keywords = "Optical flow, Variational methods, PDE, Multi-core, OpenMP", month = "March", number = "1", organization = "Centro de Tecnolog{\'i}as de la Imagen", title = "{P}arallel {I}mplementation of a {R}obust {O}ptical {F}low {T}echnique", url = "http://www.ctim.es/research_works/parallel_robust_optical_flow/", year = "2012", }

---

Resumen:

The accuracy and performance of current variational optical flow methods have considerably increased during the last years. The complexity of these techniques is high and enough care has to be taken for the implementation. The aim of this work is to present a comprehensible implementation of recent variational optical flow methods. We start with an energy model that relies on brightness and gradient constancy terms and a flow-based smoothness term. We minimize this energy model and derive an efficient implicit numerical scheme. In the experimental results, we evaluate the accuracy and performance of this implementation with the Middlebury benchmark database. We show that it is a competitive solution with respect to current methods in the literature. In order to increase the performance, we use a simple strategy to parallelize the execution on multi-core processors.

---

Texto completo:

Texto completo