The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Kertas kerja ini membincangkan resolusi super (SR) imej tunggal berdasarkan rangkaian saraf konvolusi (CNN). Adalah diketahui bahawa pemulihan komponen frekuensi tinggi dalam imej SR keluaran CNN yang dipelajari oleh ralat kuasa dua terkecil atau ralat mutlak terkecil adalah tidak mencukupi. Untuk menjana komponen frekuensi tinggi yang realistik, kaedah SR menggunakan rangkaian adversarial generatif (GAN), terdiri daripada satu penjana dan satu diskriminasi, dibangunkan. Walau bagaimanapun, apabila penjana cuba untuk mendorong salah penilaian diskriminasi, bukan sahaja komponen frekuensi tinggi yang realistik tetapi juga beberapa artifak dijana, dan indeks objektif seperti PSNR menurun. Untuk mengurangkan artifak dalam kaedah SR berasaskan GAN, kami mempertimbangkan set semua imej SR yang ralat segi empat sama antara hasil penskalaan bawah dan imej input berada dalam julat tertentu, dan mencadangkan untuk menggunakan unjuran metrik ke atas ini set konsisten dalam lapisan keluaran penjana. Teknik yang dicadangkan menjamin ketekalan antara imej SR keluaran dan imej input, dan penjana dengan unjuran yang dicadangkan boleh menjana komponen frekuensi tinggi dengan sedikit artifak sambil mengekalkan yang berfrekuensi rendah mengikut kesesuaian untuk tahap hingar yang diketahui. Percubaan berangka menunjukkan bahawa teknik yang dicadangkan mengurangkan artifak yang disertakan dalam imej SR asal kaedah SR berasaskan GAN sambil menjana komponen frekuensi tinggi yang realistik dengan nilai PSNR yang lebih baik dalam kedua-dua bebas bunyi and bising situasi. Memandangkan teknik yang dicadangkan boleh diintegrasikan ke dalam pelbagai penjana jika proses penurunan skala diketahui, kami boleh memberikan ketekalan kepada kaedah sedia ada dengan imej input tanpa merendahkan prestasi SR yang lain.
Hiroya YAMAMOTO
Ritsumeikan University
Daichi KITAHARA
Ritsumeikan University
Hiroki KURODA
Ritsumeikan University
Akira HIRABAYASHI
Ritsumeikan 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.
Salinan
Hiroya YAMAMOTO, Daichi KITAHARA, Hiroki KURODA, Akira HIRABAYASHI, "Image Super-Resolution via Generative Adversarial Networks Using Metric Projections onto Consistent Sets for Low-Resolution Inputs" in IEICE TRANSACTIONS on Fundamentals,
vol. E105-A, no. 4, pp. 704-718, April 2022, doi: 10.1587/transfun.2021EAP1038.
Abstract: This paper addresses single image super-resolution (SR) based on convolutional neural networks (CNNs). It is known that recovery of high-frequency components in output SR images of CNNs learned by the least square errors or least absolute errors is insufficient. To generate realistic high-frequency components, SR methods using generative adversarial networks (GANs), composed of one generator and one discriminator, are developed. However, when the generator tries to induce the discriminator's misjudgment, not only realistic high-frequency components but also some artifacts are generated, and objective indices such as PSNR decrease. To reduce the artifacts in the GAN-based SR methods, we consider the set of all SR images whose square errors between downscaling results and the input image are within a certain range, and propose to apply the metric projection onto this consistent set in the output layers of the generators. The proposed technique guarantees the consistency between output SR images and input images, and the generators with the proposed projection can generate high-frequency components with few artifacts while keeping low-frequency ones as appropriate for the known noise level. Numerical experiments show that the proposed technique reduces artifacts included in the original SR images of a GAN-based SR method while generating realistic high-frequency components with better PSNR values in both noise-free and noisy situations. Since the proposed technique can be integrated into various generators if the downscaling process is known, we can give the consistency to existing methods with the input images without degrading other SR performance.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/transfun.2021EAP1038/_p
Salinan
@ARTICLE{e105-a_4_704,
author={Hiroya YAMAMOTO, Daichi KITAHARA, Hiroki KURODA, Akira HIRABAYASHI, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Image Super-Resolution via Generative Adversarial Networks Using Metric Projections onto Consistent Sets for Low-Resolution Inputs},
year={2022},
volume={E105-A},
number={4},
pages={704-718},
abstract={This paper addresses single image super-resolution (SR) based on convolutional neural networks (CNNs). It is known that recovery of high-frequency components in output SR images of CNNs learned by the least square errors or least absolute errors is insufficient. To generate realistic high-frequency components, SR methods using generative adversarial networks (GANs), composed of one generator and one discriminator, are developed. However, when the generator tries to induce the discriminator's misjudgment, not only realistic high-frequency components but also some artifacts are generated, and objective indices such as PSNR decrease. To reduce the artifacts in the GAN-based SR methods, we consider the set of all SR images whose square errors between downscaling results and the input image are within a certain range, and propose to apply the metric projection onto this consistent set in the output layers of the generators. The proposed technique guarantees the consistency between output SR images and input images, and the generators with the proposed projection can generate high-frequency components with few artifacts while keeping low-frequency ones as appropriate for the known noise level. Numerical experiments show that the proposed technique reduces artifacts included in the original SR images of a GAN-based SR method while generating realistic high-frequency components with better PSNR values in both noise-free and noisy situations. Since the proposed technique can be integrated into various generators if the downscaling process is known, we can give the consistency to existing methods with the input images without degrading other SR performance.},
keywords={},
doi={10.1587/transfun.2021EAP1038},
ISSN={1745-1337},
month={April},}
Salinan
TY - JOUR
TI - Image Super-Resolution via Generative Adversarial Networks Using Metric Projections onto Consistent Sets for Low-Resolution Inputs
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 704
EP - 718
AU - Hiroya YAMAMOTO
AU - Daichi KITAHARA
AU - Hiroki KURODA
AU - Akira HIRABAYASHI
PY - 2022
DO - 10.1587/transfun.2021EAP1038
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E105-A
IS - 4
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - April 2022
AB - This paper addresses single image super-resolution (SR) based on convolutional neural networks (CNNs). It is known that recovery of high-frequency components in output SR images of CNNs learned by the least square errors or least absolute errors is insufficient. To generate realistic high-frequency components, SR methods using generative adversarial networks (GANs), composed of one generator and one discriminator, are developed. However, when the generator tries to induce the discriminator's misjudgment, not only realistic high-frequency components but also some artifacts are generated, and objective indices such as PSNR decrease. To reduce the artifacts in the GAN-based SR methods, we consider the set of all SR images whose square errors between downscaling results and the input image are within a certain range, and propose to apply the metric projection onto this consistent set in the output layers of the generators. The proposed technique guarantees the consistency between output SR images and input images, and the generators with the proposed projection can generate high-frequency components with few artifacts while keeping low-frequency ones as appropriate for the known noise level. Numerical experiments show that the proposed technique reduces artifacts included in the original SR images of a GAN-based SR method while generating realistic high-frequency components with better PSNR values in both noise-free and noisy situations. Since the proposed technique can be integrated into various generators if the downscaling process is known, we can give the consistency to existing methods with the input images without degrading other SR performance.
ER -