Back to Search
Start Over
Fast and accurate superpixel algorithms for 360[formula omitted] images.
- Source :
-
Signal Processing . Dec2021, Vol. 189, pN.PAG-N.PAG. 1p. - Publication Year :
- 2021
-
Abstract
- • We propose a method for multiscale, border-guided 360 ∘ image oversegmentation. • We propose a superpixel method for 360 ∘ images that concern borders and regularity. • We propose an annotation-free metric for assessing 360 ∘ image segmentation methods. • The novel methods are compared and assessed under reference metrics and datasets. • Our methods achieve state-of-the-art results and are faster than peering approaches. Superpixels are fundamental in many visual computing applications, and most existing algorithms are designed to work with pinhole-based images. However, immersive applications are gaining visibility with the growing number of devices for capturing and visualizing 360 ∘ media. This paper introduces two fast and accurate superpixel algorithms tailored to the spherical domain. The methods are assessed under common figures of merit, and a benchmark annotated dataset. Additionally, an annotation-free evaluation metric for exploiting larger datasets is introduced. The methods introduced in this paper perform quantitatively close to or better than state-of-the-art approaches and are faster. [ABSTRACT FROM AUTHOR]
- Subjects :
- *PIXELS
*ALGORITHMS
*IMAGE segmentation
Subjects
Details
- Language :
- English
- ISSN :
- 01651684
- Volume :
- 189
- Database :
- Academic Search Index
- Journal :
- Signal Processing
- Publication Type :
- Academic Journal
- Accession number :
- 152427032
- Full Text :
- https://doi.org/10.1016/j.sigpro.2021.108277