Disparity Estimation for Image Fusion in a Multi-aperture Camera. Mustaniemi, J.; Kannala, J.; and Heikkilä, J. Volume 9257. Computer Analysis of Images and Patterns. CAIP 2015. Lecture Notes in Computer Science, pages 158-170. Springer, Cham, 2015.
Computer Analysis of Images and Patterns. CAIP 2015. Lecture Notes in Computer Science [link]Website  abstract   bibtex   
In this paper, an image fusion algorithm is proposed for a multi-aperture camera. Such camera is a worthy alternative to traditional Bayer filter camera in terms of image quality, camera size and camera features. The camera consists of several camera units, each having ded- icated optics and color filter. The main challenge of a multi-aperture camera arises from the fact that each camera unit has a slightly differ- ent viewpoint. Our image fusion algorithm corrects the parallax error between the sub-images using a disparity map, which is estimated from the multi-spectral images.We improve the disparity estimation by combining matching costs over multiple views with help of trifocal tensors. Images are matched using two alternative matching costs, mutual information andCensus transform.We also compare two different disparity estimation methods, graph cuts and semi-globalmatching. The results show that the overall quality of the fused images is near the reference images.
@inbook{
 type = {inbook},
 year = {2015},
 identifiers = {[object Object]},
 keywords = {Census transform,Mutual information,Trifocal tensor},
 pages = {158-170},
 volume = {9257},
 websites = {http://link.springer.com/10.1007/978-3-319-23117-4_14},
 publisher = {Springer, Cham},
 id = {3d2150d6-7dfe-39dc-bcae-02336bbe939f},
 created = {2019-09-15T16:34:27.633Z},
 file_attached = {false},
 profile_id = {bddcf02d-403b-3b06-9def-6d15cc293e20},
 last_modified = {2019-09-23T18:20:08.124Z},
 read = {true},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 source_type = {CONF},
 private_publication = {false},
 abstract = {In this paper, an image fusion algorithm is proposed for a multi-aperture camera. Such camera is a worthy alternative to traditional Bayer filter camera in terms of image quality, camera size and camera features. The camera consists of several camera units, each having ded- icated optics and color filter. The main challenge of a multi-aperture camera arises from the fact that each camera unit has a slightly differ- ent viewpoint. Our image fusion algorithm corrects the parallax error between the sub-images using a disparity map, which is estimated from the multi-spectral images.We improve the disparity estimation by combining matching costs over multiple views with help of trifocal tensors. Images are matched using two alternative matching costs, mutual information andCensus transform.We also compare two different disparity estimation methods, graph cuts and semi-globalmatching. The results show that the overall quality of the fused images is near the reference images.},
 bibtype = {inbook},
 author = {Mustaniemi, Janne and Kannala, Juho and Heikkilä, Janne},
 chapter = {Disparity Estimation for Image Fusion in a Multi-aperture Camera},
 title = {Computer Analysis of Images and Patterns. CAIP 2015. Lecture Notes in Computer Science}
}
Downloads: 0