Deep network for simultaneous stereo matching and dehazing

Taeyong Song, Youngjung Kim, Changjae Oh, Kwanghoon Sohn

Research output: Contribution to conferencePaper

1 Citation (Scopus)

Abstract

Unveiling the image structure and dense correspondence under the haze layer remains a challenging task, since the scattering effects cause image features to be less distinctive. In this paper, we introduce a deep network that simultaneously estimates a clear latent image and disparity from a hazy stereo image pair. To this end, inspired by a physical model of hazy image acquisition, we propose a dehazing loss function which serves as an additional cue for establishing dense correspondence. We show that stereo matching and dehazing can be synergistically formulated by incorporating depth information from haze transmission into the stereo matching process, and vice versa. As a result, our method estimates high-quality disparity for scenes in scattering media, and produces appearance images with enhanced visibility. We quantitatively evaluate the proposed method on synthetic datasets and provide an extensive ablation study. Experimental results demonstrate that our approach outperforms the recent state-of-the-art methods on both dehazing and stereo matching tasks.

Original languageEnglish
Publication statusPublished - 2019 Jan 1
Event29th British Machine Vision Conference, BMVC 2018 - Newcastle, United Kingdom
Duration: 2018 Sep 32018 Sep 6

Conference

Conference29th British Machine Vision Conference, BMVC 2018
CountryUnited Kingdom
CityNewcastle
Period18/9/318/9/6

Fingerprint

Scattering
Image acquisition
Ablation
Visibility

All Science Journal Classification (ASJC) codes

  • Computer Vision and Pattern Recognition

Cite this

Song, T., Kim, Y., Oh, C., & Sohn, K. (2019). Deep network for simultaneous stereo matching and dehazing. Paper presented at 29th British Machine Vision Conference, BMVC 2018, Newcastle, United Kingdom.
Song, Taeyong ; Kim, Youngjung ; Oh, Changjae ; Sohn, Kwanghoon. / Deep network for simultaneous stereo matching and dehazing. Paper presented at 29th British Machine Vision Conference, BMVC 2018, Newcastle, United Kingdom.
@conference{31ccb0b96a414286a70d8721794644c0,
title = "Deep network for simultaneous stereo matching and dehazing",
abstract = "Unveiling the image structure and dense correspondence under the haze layer remains a challenging task, since the scattering effects cause image features to be less distinctive. In this paper, we introduce a deep network that simultaneously estimates a clear latent image and disparity from a hazy stereo image pair. To this end, inspired by a physical model of hazy image acquisition, we propose a dehazing loss function which serves as an additional cue for establishing dense correspondence. We show that stereo matching and dehazing can be synergistically formulated by incorporating depth information from haze transmission into the stereo matching process, and vice versa. As a result, our method estimates high-quality disparity for scenes in scattering media, and produces appearance images with enhanced visibility. We quantitatively evaluate the proposed method on synthetic datasets and provide an extensive ablation study. Experimental results demonstrate that our approach outperforms the recent state-of-the-art methods on both dehazing and stereo matching tasks.",
author = "Taeyong Song and Youngjung Kim and Changjae Oh and Kwanghoon Sohn",
year = "2019",
month = "1",
day = "1",
language = "English",
note = "29th British Machine Vision Conference, BMVC 2018 ; Conference date: 03-09-2018 Through 06-09-2018",

}

Song, T, Kim, Y, Oh, C & Sohn, K 2019, 'Deep network for simultaneous stereo matching and dehazing', Paper presented at 29th British Machine Vision Conference, BMVC 2018, Newcastle, United Kingdom, 18/9/3 - 18/9/6.

Deep network for simultaneous stereo matching and dehazing. / Song, Taeyong; Kim, Youngjung; Oh, Changjae; Sohn, Kwanghoon.

2019. Paper presented at 29th British Machine Vision Conference, BMVC 2018, Newcastle, United Kingdom.

Research output: Contribution to conferencePaper

TY - CONF

T1 - Deep network for simultaneous stereo matching and dehazing

AU - Song, Taeyong

AU - Kim, Youngjung

AU - Oh, Changjae

AU - Sohn, Kwanghoon

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Unveiling the image structure and dense correspondence under the haze layer remains a challenging task, since the scattering effects cause image features to be less distinctive. In this paper, we introduce a deep network that simultaneously estimates a clear latent image and disparity from a hazy stereo image pair. To this end, inspired by a physical model of hazy image acquisition, we propose a dehazing loss function which serves as an additional cue for establishing dense correspondence. We show that stereo matching and dehazing can be synergistically formulated by incorporating depth information from haze transmission into the stereo matching process, and vice versa. As a result, our method estimates high-quality disparity for scenes in scattering media, and produces appearance images with enhanced visibility. We quantitatively evaluate the proposed method on synthetic datasets and provide an extensive ablation study. Experimental results demonstrate that our approach outperforms the recent state-of-the-art methods on both dehazing and stereo matching tasks.

AB - Unveiling the image structure and dense correspondence under the haze layer remains a challenging task, since the scattering effects cause image features to be less distinctive. In this paper, we introduce a deep network that simultaneously estimates a clear latent image and disparity from a hazy stereo image pair. To this end, inspired by a physical model of hazy image acquisition, we propose a dehazing loss function which serves as an additional cue for establishing dense correspondence. We show that stereo matching and dehazing can be synergistically formulated by incorporating depth information from haze transmission into the stereo matching process, and vice versa. As a result, our method estimates high-quality disparity for scenes in scattering media, and produces appearance images with enhanced visibility. We quantitatively evaluate the proposed method on synthetic datasets and provide an extensive ablation study. Experimental results demonstrate that our approach outperforms the recent state-of-the-art methods on both dehazing and stereo matching tasks.

UR - http://www.scopus.com/inward/record.url?scp=85072312755&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85072312755&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:85072312755

ER -

Song T, Kim Y, Oh C, Sohn K. Deep network for simultaneous stereo matching and dehazing. 2019. Paper presented at 29th British Machine Vision Conference, BMVC 2018, Newcastle, United Kingdom.