Dual branches network for image super‐resolution
نویسندگان
چکیده
منابع مشابه
Preconditioners for regularized image superresolution
Superresolution reconstruction produces a high resolution image from a set of low resolution images. Previous work on superresolution [3, 6, 10, 12] had not adequately addressed the computational issues for this problem. In this paper, we propose efficient block circulant preconditioners for solving the regularized superresolution problem by CG. Effectiveness of our preconditioners is demonstra...
متن کاملGraphcut Texture Synthesis for Single-Image Superresolution
Texture synthesis has proven successful at imitating a wide variety of textures. Adding additional constraints (in the form of a low-resolution version of the texture to be synthesized) makes it possible to use texture synthesis methods for texture superresolution. The Single-image Superresolution Problem The problem we are trying to solve is the following: Given: a low-resolution image and hig...
متن کاملDual Z-Source Network Dual-Input Dual-Output Inverter
This paper presents a modified nine switch inverter with two inputs and two Z-source networks. This inverter has two DC inputs and two AC outputs. Input DC voltages can be boosted to the required level. Amplitude, frequency and phase of AC output voltages can be controlled, independently. The proposed converter can be used in applications with two unregulated DC sources, which require feeding ...
متن کاملA computationally efficient superresolution image reconstruction algorithm
Superresolution reconstruction produces a high-resolution image from a set of low-resolution images. Previous iterative methods for superresolution had not adequately addressed the computational and numerical issues for this ill-conditioned and typically underdetermined large scale problem. We propose efficient block circulant preconditioners for solving the Tikhonov-regularized superresolution...
متن کاملDual Recovery Network with Online Compensation for Image Super-Resolution
The image super-resolution (SR) methods will essentially lead to a loss of some high-frequency (HF) information when predicting high-resolution (HR) images from low-resolution (LR) images without using external references. To address that, we additionally utilize online retrieved data to facilitate image SR in a unified deep framework. A novel dual highfrequency recovery network (DHN) is propos...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Electronics Letters
سال: 2019
ISSN: 0013-5194,1350-911X
DOI: 10.1049/el.2018.7562