A learned joint depth and intensity prior using Markov Random fields

Abstract : We present a joint prior that takes intensity and depth information into account. The prior is defined using a flexible Field-of-Experts model and is learned from a database of natural images. It is a generative model and has an efficient method for sampling. We use sampling from the model to perform in painting and up sampling of depth maps when intensity information is available. We show that including the intensity information in the prior improves the results obtained from the model. We also compare to another two-channel inpainting approach and show superior results.
Document type :
Conference papers
Complete list of metadatas

Cited literature [18 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-00880486
Contributor : Peter Sturm <>
Submitted on : Wednesday, November 6, 2013 - 11:27:18 AM
Last modification on : Wednesday, April 11, 2018 - 1:59:45 AM
Long-term archiving on : Friday, February 7, 2014 - 6:20:21 AM

File

dfoe_camera_ready.pdf
Files produced by the author(s)

Identifiers

Collections

Citation

Daniel Herrera Castro, Juho Kannala, Peter Sturm, Janne Heikkilä. A learned joint depth and intensity prior using Markov Random fields. 3DV 2013 - International Conference on 3D Vision, Jun 2013, Seattle, United States. pp.17-24, ⟨10.1109/3DV.2013.11⟩. ⟨hal-00880486⟩

Share

Metrics

Record views

581

Files downloads

595