Optimal Geometric Matching for Patch-Based Object Detection

We present an efficient method to determine the optimal matching of two patch-based image object representations under rotation, scaling, and translation (RST). This use of patches is equivalent to a fullyconnected part-based model, for which the presented approach offers an efficient procedure to d...

Descripción completa

Detalles Bibliográficos
Autores: Keysers, Daniel, Deselaers, Thomas, Breuel, Thomas M.
Tipo de recurso: artículo
Fecha de publicación:2007
País:España
Institución:Universitat Autònoma de Barcelona
Repositorio:Dipòsit Digital de Documents de la UAB
Idioma:inglés
OAI Identifier:oai:ddd.uab.cat:24579
Acceso en línea:https://ddd.uab.cat/record/24579
https://dx.doi.org/urn:doi:10.5565/rev/elcvia.136
Access Level:acceso abierto
Palabra clave:Object recognition
Registration and matching
Reconeixement objecte
Registre i rastreig
Reconocimiento objeto
Registro y rastreo
Descripción
Sumario:We present an efficient method to determine the optimal matching of two patch-based image object representations under rotation, scaling, and translation (RST). This use of patches is equivalent to a fullyconnected part-based model, for which the presented approach offers an efficient procedure to determine the best fit. While other approaches that use fully connected models have a high complexity in the number of parts used, we achieve linear complexity in that variable, because we only allow RST-matchings. The presented approach is used for object recognition in images: by matching images that contain certain objects to a test image, we can detect whether the test image contains an object of that class or not. We evaluate this approach on the Caltech data and obtain very competitive results.