Feature Forwarding for Efficient Single Image Dehazing
P. Morales, T. Klinghoffer, and S. Lee. (2019)cite arxiv:1904.09059Comment: Accepted to the NTIRE 2019 CVPR Workshop. Paper number 77. 8 Pages.
Abstract
Haze degrades content and obscures information of images, which can
negatively impact vision-based decision-making in real-time systems. In this
paper, we propose an efficient fully convolutional neural network (CNN) image
dehazing method designed to run on edge graphical processing units (GPUs). We
utilize three variants of our architecture to explore the dependency of dehazed
image quality on parameter count and model design. The first two variants
presented, a small and big version, make use of a single efficient
encoder-decoder convolutional feature extractor. The final variant utilizes a
pair of encoder-decoders for atmospheric light and transmission map estimation.
Each variant ends with an image refinement pyramid pooling network to form the
final dehazed image. For the big variant of the single-encoder network, we
demonstrate state-of-the-art performance on the NYU Depth dataset. For the
small variant, we maintain competitive performance on the super-resolution
O/I-HAZE datasets without the need for image cropping. Finally, we examine some
challenges presented by the Dense-Haze dataset when leveraging CNN
architectures for dehazing of dense haze imagery and examine the impact of loss
function selection on image quality. Benchmarks are included to show the
feasibility of introducing this approach into real-time systems.
Description
Feature Forwarding for Efficient Single Image Dehazing
%0 Generic
%1 morales2019feature
%A Morales, Peter
%A Klinghoffer, Tzofi
%A Lee, Seung Jae
%D 2019
%K myown
%T Feature Forwarding for Efficient Single Image Dehazing
%U http://arxiv.org/abs/1904.09059
%X Haze degrades content and obscures information of images, which can
negatively impact vision-based decision-making in real-time systems. In this
paper, we propose an efficient fully convolutional neural network (CNN) image
dehazing method designed to run on edge graphical processing units (GPUs). We
utilize three variants of our architecture to explore the dependency of dehazed
image quality on parameter count and model design. The first two variants
presented, a small and big version, make use of a single efficient
encoder-decoder convolutional feature extractor. The final variant utilizes a
pair of encoder-decoders for atmospheric light and transmission map estimation.
Each variant ends with an image refinement pyramid pooling network to form the
final dehazed image. For the big variant of the single-encoder network, we
demonstrate state-of-the-art performance on the NYU Depth dataset. For the
small variant, we maintain competitive performance on the super-resolution
O/I-HAZE datasets without the need for image cropping. Finally, we examine some
challenges presented by the Dense-Haze dataset when leveraging CNN
architectures for dehazing of dense haze imagery and examine the impact of loss
function selection on image quality. Benchmarks are included to show the
feasibility of introducing this approach into real-time systems.
@misc{morales2019feature,
abstract = {Haze degrades content and obscures information of images, which can
negatively impact vision-based decision-making in real-time systems. In this
paper, we propose an efficient fully convolutional neural network (CNN) image
dehazing method designed to run on edge graphical processing units (GPUs). We
utilize three variants of our architecture to explore the dependency of dehazed
image quality on parameter count and model design. The first two variants
presented, a small and big version, make use of a single efficient
encoder-decoder convolutional feature extractor. The final variant utilizes a
pair of encoder-decoders for atmospheric light and transmission map estimation.
Each variant ends with an image refinement pyramid pooling network to form the
final dehazed image. For the big variant of the single-encoder network, we
demonstrate state-of-the-art performance on the NYU Depth dataset. For the
small variant, we maintain competitive performance on the super-resolution
O/I-HAZE datasets without the need for image cropping. Finally, we examine some
challenges presented by the Dense-Haze dataset when leveraging CNN
architectures for dehazing of dense haze imagery and examine the impact of loss
function selection on image quality. Benchmarks are included to show the
feasibility of introducing this approach into real-time systems.},
added-at = {2019-10-18T19:59:43.000+0200},
author = {Morales, Peter and Klinghoffer, Tzofi and Lee, Seung Jae},
biburl = {https://www.bibsonomy.org/bibtex/2b4024c94169d2093ff2b49b17af2d0e4/pmorales},
description = {Feature Forwarding for Efficient Single Image Dehazing},
interhash = {42b457621bef1f50df3475d9e55c6825},
intrahash = {b4024c94169d2093ff2b49b17af2d0e4},
keywords = {myown},
note = {cite arxiv:1904.09059Comment: Accepted to the NTIRE 2019 CVPR Workshop. Paper number 77. 8 Pages},
timestamp = {2019-10-18T19:59:43.000+0200},
title = {Feature Forwarding for Efficient Single Image Dehazing},
url = {http://arxiv.org/abs/1904.09059},
year = 2019
}