In this letter, we propose a novel method for face hallucination by learning a new distance metric in the low-resolution (LR) patch space (source space). Local patch-based face hallucination methods usually assume that the two manifolds formed by LR and high-resolution (HR) image patches have similar local geometry. However, this assumption does not hold well in practice. Motivated by metric learning in machine learning, we propose to learn a new distance metric in the source space, under the supervision of the true local geometry in the target space (HR patch space). The learned new metric gives more freedom to the presentation of local geometry in the source space, and thus the local geometries of source and target space turn to be more consistent. Experiments conducted on two datasets demonstrate that the proposed method is superior to the state-of-the-art face hallucination and image super-resolution (SR) methods.
Yuanpeng ZOU
Tsinghua University
Fei ZHOU
Tsinghua University
Qingmin LIAO
Tsinghua University
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Yuanpeng ZOU, Fei ZHOU, Qingmin LIAO, "Face Hallucination by Learning Local Distance Metric" in IEICE TRANSACTIONS on Information,
vol. E100-D, no. 2, pp. 384-387, February 2017, doi: 10.1587/transinf.2016EDL8200.
Abstract: In this letter, we propose a novel method for face hallucination by learning a new distance metric in the low-resolution (LR) patch space (source space). Local patch-based face hallucination methods usually assume that the two manifolds formed by LR and high-resolution (HR) image patches have similar local geometry. However, this assumption does not hold well in practice. Motivated by metric learning in machine learning, we propose to learn a new distance metric in the source space, under the supervision of the true local geometry in the target space (HR patch space). The learned new metric gives more freedom to the presentation of local geometry in the source space, and thus the local geometries of source and target space turn to be more consistent. Experiments conducted on two datasets demonstrate that the proposed method is superior to the state-of-the-art face hallucination and image super-resolution (SR) methods.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2016EDL8200/_p
Copy
@ARTICLE{e100-d_2_384,
author={Yuanpeng ZOU, Fei ZHOU, Qingmin LIAO, },
journal={IEICE TRANSACTIONS on Information},
title={Face Hallucination by Learning Local Distance Metric},
year={2017},
volume={E100-D},
number={2},
pages={384-387},
abstract={In this letter, we propose a novel method for face hallucination by learning a new distance metric in the low-resolution (LR) patch space (source space). Local patch-based face hallucination methods usually assume that the two manifolds formed by LR and high-resolution (HR) image patches have similar local geometry. However, this assumption does not hold well in practice. Motivated by metric learning in machine learning, we propose to learn a new distance metric in the source space, under the supervision of the true local geometry in the target space (HR patch space). The learned new metric gives more freedom to the presentation of local geometry in the source space, and thus the local geometries of source and target space turn to be more consistent. Experiments conducted on two datasets demonstrate that the proposed method is superior to the state-of-the-art face hallucination and image super-resolution (SR) methods.},
keywords={},
doi={10.1587/transinf.2016EDL8200},
ISSN={1745-1361},
month={February},}
Copy
TY - JOUR
TI - Face Hallucination by Learning Local Distance Metric
T2 - IEICE TRANSACTIONS on Information
SP - 384
EP - 387
AU - Yuanpeng ZOU
AU - Fei ZHOU
AU - Qingmin LIAO
PY - 2017
DO - 10.1587/transinf.2016EDL8200
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E100-D
IS - 2
JA - IEICE TRANSACTIONS on Information
Y1 - February 2017
AB - In this letter, we propose a novel method for face hallucination by learning a new distance metric in the low-resolution (LR) patch space (source space). Local patch-based face hallucination methods usually assume that the two manifolds formed by LR and high-resolution (HR) image patches have similar local geometry. However, this assumption does not hold well in practice. Motivated by metric learning in machine learning, we propose to learn a new distance metric in the source space, under the supervision of the true local geometry in the target space (HR patch space). The learned new metric gives more freedom to the presentation of local geometry in the source space, and thus the local geometries of source and target space turn to be more consistent. Experiments conducted on two datasets demonstrate that the proposed method is superior to the state-of-the-art face hallucination and image super-resolution (SR) methods.
ER -