Special Issue on Deep Learning for Image Super-Resolution
摘要截稿:
全文截稿: 2018-08-31
影响因子: 4.438
期刊难度:
CCF分类: C类
中科院JCR分区:
• 大类 : 计算机科学 - 2区
• 小类 : 计算机:人工智能 - 2区
Overview
The goal of image super-resolution (SR) is to restore a visually pleasing high-resolution (HR) image from a low-resolution (LR) input image or video sequence. HR images have higher pixel densities and finer details than LR images. Image SR has been proved to be of great significance in many applications, such as video surveillance, ultra-high definition TV, low-resolution face recognition and remote sensing imaging. Benefiting from its broad application prospects, SR has attracted huge interest, and currently is one of the most active research topics in image processing and computer vision.
Early interpolation-based image SR methods are extremely simple and fast. Unfortunately, severe aliasing and blurring effects make interpolation-based SR suboptimal in restoring fine texture details. Reconstruction-based image SR methods combine elaborately designed image prior models with reconstruction constraints, and can restore fine structures. However, these image priors usually are incapable of modeling complex and varying contexts of natural image.
In the past decade, most researches focus on learning-based image SR. It utilizes machine learning techniques to capture the relationships between LR image patches and their HR counterparts from some samples. Recently due to fast advances in deep learning, deep network-based SR has shown promising performance in certain applications. However, there are still many challenging open topics of deep learning for image SR, e.g. new objective functions, new architectures, large scale images, depth images, various types of corruption, and new applications.
Therefore, this special issue emphasizes the important role of deep learning for image SR. It aims to call for the state-of-the-art researches in the theory, algorithm, modeling, system and application of deep learning-based SR and to demonstrate the latest efforts of relevant researchers.
The list of possible topics includes, but is not limited to:
- Review/survey/vision of deep learning for SR
- New image databases for deep learning for SR
- New objective functions of deep learning for SR
- New deep network architectures for SR
- Combining deep learning with traditional SR approaches
- Combining deep learning with image priors
- Deep learning for large scale SR
- Deep learning for SR with different or unknown types of corruption
- Deep learning for video sequence SR
- Deep learning for SR for special types of images
- Deep learning for depth image SR
- Hybrid RGB and depth image SR with deep learning
- Acceleration of deep learning for SR
- Hardware and systems of deep learning for SR
- Deep learning-based SR applications in video surveillance, ultra-high definition TV, face hallucination, biometrics, medical imaging, remote sensing, LR face recognition, etc.