View-Consistent 4D Light Field Superpixel Segmentation

Numair Khan, Qian Zhang, Lucas Kasser, Henry Stone, Min H. Kim, James Tompkin; The IEEE International Conference on Computer Vision (ICCV), 2019, pp. 7811-7819

Abstract


Many 4D light field processing applications rely on superpixel segmentations, for which occlusion-aware view consistency is important. Yet, existing methods often enforce consistency by propagating clusters from a central view only, which can lead to inconsistent superpixels for non-central views. Our proposed approach combines an occlusion-aware angular segmentation in horizontal and vertical EPI spaces with an occlusion-aware clustering and propagation step across all views. Qualitative video demonstrations show that this helps to remove flickering and inconsistent boundary shapes versus the state-of-the-art approach, and quantitative metrics reflect these findings with improved boundary accuracy and view consistency scores.

Related Material


[pdf] [supp]
[bibtex]
@InProceedings{Khan_2019_ICCV,
author = {Khan, Numair and Zhang, Qian and Kasser, Lucas and Stone, Henry and Kim, Min H. and Tompkin, James},
title = {View-Consistent 4D Light Field Superpixel Segmentation},
booktitle = {The IEEE International Conference on Computer Vision (ICCV)},
month = {October},
year = {2019}
}