U-shaped Networks for Shape from Light Field

Stefan Heber, Wei Yu, Thomas Pock

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

Abstract

This paper presents a novel technique for Shape from Light Field (SfLF), that utilizes deep learning strategies. Our model is based on a fully convolutional network, that involves two symmetric parts, an encoding and a decoding part, leading to a u-shaped network architecture. By leveraging a recently proposed Light Field (LF) dataset, we are able to effectively train our model using supervised training. To process an entire LF we split the LF data into the corresponding Epipolar Plane Image (EPI) representation and predict each EPI separately. This strategy provides good reconstruction results combined with a fast prediction time. In the experimental section we compare our method to the state of the art. The method performs well in terms of depth accuracy, and is able to outperform competing methods in terms of prediction time by a large margin.
Original languageEnglish
Title of host publicationBritish Machine Vision Conference, BMVC 2016
Publication statusPublished - Sep 2016

Fingerprint

Network architecture
Decoding
Deep learning

Cite this

Heber, S., Yu, W., & Pock, T. (2016). U-shaped Networks for Shape from Light Field. In British Machine Vision Conference, BMVC 2016

U-shaped Networks for Shape from Light Field. / Heber, Stefan; Yu, Wei; Pock, Thomas.

British Machine Vision Conference, BMVC 2016. 2016.

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

Heber, S, Yu, W & Pock, T 2016, U-shaped Networks for Shape from Light Field. in British Machine Vision Conference, BMVC 2016.
Heber S, Yu W, Pock T. U-shaped Networks for Shape from Light Field. In British Machine Vision Conference, BMVC 2016. 2016
Heber, Stefan ; Yu, Wei ; Pock, Thomas. / U-shaped Networks for Shape from Light Field. British Machine Vision Conference, BMVC 2016. 2016.
@inproceedings{79165723856c4de3b1457cba74e93fb5,
title = "U-shaped Networks for Shape from Light Field",
abstract = "This paper presents a novel technique for Shape from Light Field (SfLF), that utilizes deep learning strategies. Our model is based on a fully convolutional network, that involves two symmetric parts, an encoding and a decoding part, leading to a u-shaped network architecture. By leveraging a recently proposed Light Field (LF) dataset, we are able to effectively train our model using supervised training. To process an entire LF we split the LF data into the corresponding Epipolar Plane Image (EPI) representation and predict each EPI separately. This strategy provides good reconstruction results combined with a fast prediction time. In the experimental section we compare our method to the state of the art. The method performs well in terms of depth accuracy, and is able to outperform competing methods in terms of prediction time by a large margin.",
author = "Stefan Heber and Wei Yu and Thomas Pock",
year = "2016",
month = "9",
language = "English",
booktitle = "British Machine Vision Conference, BMVC 2016",

}

TY - GEN

T1 - U-shaped Networks for Shape from Light Field

AU - Heber, Stefan

AU - Yu, Wei

AU - Pock, Thomas

PY - 2016/9

Y1 - 2016/9

N2 - This paper presents a novel technique for Shape from Light Field (SfLF), that utilizes deep learning strategies. Our model is based on a fully convolutional network, that involves two symmetric parts, an encoding and a decoding part, leading to a u-shaped network architecture. By leveraging a recently proposed Light Field (LF) dataset, we are able to effectively train our model using supervised training. To process an entire LF we split the LF data into the corresponding Epipolar Plane Image (EPI) representation and predict each EPI separately. This strategy provides good reconstruction results combined with a fast prediction time. In the experimental section we compare our method to the state of the art. The method performs well in terms of depth accuracy, and is able to outperform competing methods in terms of prediction time by a large margin.

AB - This paper presents a novel technique for Shape from Light Field (SfLF), that utilizes deep learning strategies. Our model is based on a fully convolutional network, that involves two symmetric parts, an encoding and a decoding part, leading to a u-shaped network architecture. By leveraging a recently proposed Light Field (LF) dataset, we are able to effectively train our model using supervised training. To process an entire LF we split the LF data into the corresponding Epipolar Plane Image (EPI) representation and predict each EPI separately. This strategy provides good reconstruction results combined with a fast prediction time. In the experimental section we compare our method to the state of the art. The method performs well in terms of depth accuracy, and is able to outperform competing methods in terms of prediction time by a large margin.

M3 - Conference contribution

BT - British Machine Vision Conference, BMVC 2016

ER -