Learning on the Edge: Investigating Boundary Filters in CNNs
Convolutional neural networks (CNNs) handle the case where filters extend beyond the image boundary using several heuristics, such as zero, repeat or mean padding. These schemes are applied in an ad-hoc fashion and, being weakly related to the image content and oblivious of the target task, result i...
Saved in:
Published in | International journal of computer vision Vol. 128; no. 4; pp. 773 - 782 |
---|---|
Main Authors | , , , |
Format | Journal Article |
Language | English |
Published |
New York
Springer US
01.04.2020
Springer Springer Nature B.V |
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | Convolutional neural networks (CNNs) handle the case where filters extend beyond the image boundary using several heuristics, such as zero, repeat or mean padding. These schemes are applied in an ad-hoc fashion and, being weakly related to the image content and oblivious of the target task, result in low output quality at the boundary. In this paper, we propose a simple and effective improvement that
learns
the boundary handling itself. At training-time, the network is provided with a separate set of explicit boundary filters. At testing-time, we use these filters which have learned to extrapolate features at the boundary in an optimal way for the specific task. Our extensive evaluation, over a wide range of architectural changes (variations of layers, feature channels, or both), shows how the explicit filters result in improved boundary handling. Furthermore, we investigate the efficacy of variations of such boundary filters with respect to convergence speed and accuracy. Finally, we demonstrate an improvement of 5–20% across the board of typical CNN applications (colorization, de-Bayering, optical flow, disparity estimation, and super-resolution). Supplementary material and code can be downloaded from the project page:
http://geometry.cs.ucl.ac.uk/projects/2019/investigating-edge/
. |
---|---|
ISSN: | 0920-5691 1573-1405 |
DOI: | 10.1007/s11263-019-01223-y |