Tomographic approach for the quantitative scene reconstruction from light field images

Abstract : Current computational methods for light field photography model the ray-tracing geometry inside the plenoptic camera. This representation of the problem, and some common approximations, can lead to errors in the estimation of object sizes and positions. We propose are presentation that leads to the correct reconstruction of object sizes and distances to the camera,by showing that light field images can be interpreted as limited angle cone-beam tomography acquisitions. We then quantitatively analyze its impact on image refocusing, depth estimation and volumetric reconstructions, comparing it against other possible representations. Finally, we validate these results with numerical and real-world examples.
Type de document :
Article dans une revue
Optics Express, Optical Society of America, 2018, 26 (18), pp.22574. 〈10.1364/OE.26.022574 〉
Liste complète des métadonnées

https://hal.inria.fr/hal-02042417
Contributeur : Xavier Granier <>
Soumis le : mercredi 20 février 2019 - 13:49:27
Dernière modification le : vendredi 22 février 2019 - 01:18:41

Identifiants

Citation

Nicola Viganò, Henri Der Sarkissian, Charlotte Herzog, Ombeline De La Rochefoucauld, Robert Van Liere, et al.. Tomographic approach for the quantitative scene reconstruction from light field images. Optics Express, Optical Society of America, 2018, 26 (18), pp.22574. 〈10.1364/OE.26.022574 〉. 〈hal-02042417〉

Partager

Métriques

Consultations de la notice

18