Learning View Generalization Functions
| dc.creator | Breuel, Thomas M. | |
| dc.date | 2007-12-02 | |
| dc.date.accessioned | 2026-07-07T08:46:46Z | |
| dc.date.available | 2026-07-07T08:46:46Z | |
| dc.description | Learning object models from views in 3D visual object recognition is usually formulated either as a function approximation problem of a function describing the view-manifold of an object, or as that of learning a class-conditional density. This paper describes an alternative framework for learning in visual object recognition, that of learning the view-generalization function. Using the view-generalization function, an observer can perform Bayes-optimal 3D object recognition given one or more 2D training views directly, without the need for a separate model acquisition step. The paper shows that view generalization functions can be computationally practical by restating two widely-used methods, the eigenspace and linear combination of views approaches, in a view generalization framework. The paper relates the approach to recent methods for object recognition based on non-uniform blurring. The paper presents results both on simulated 3D ``paperclip'' objects and real-world images from the COIL-100 database showing that useful view-generalization functions can be realistically be learned from a comparatively small number of training examples. | |
| dc.identifier | https://arxiv.org/abs/0712.0136 | |
| dc.identifier | http://arxiv.org/abs/0712.0136 | |
| dc.identifier.uri | http://salesiana.dossiersoluciones.com/handle/123456789/143365 | |
| dc.subject | Computer Vision and Pattern Recognition | |
| dc.subject | I.4.8; I.2.10 | |
| dc.title | Learning View Generalization Functions | |
| dc.type | text |