Lightweight Dual-Stream Residual Network for Single Image Super-Resolution

Jiang, Yichun; Liu, Yunqing; Zhan, Weida; Zhu, Daming

doi:10.1109/access.2021.3112002

Cited by 10 publications

(4 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On the other hand, so-called lightweight SR models, aiming to reduce the number of parameters and FLOPs while maintaining performance, is actively being studied [8,9,10,11,22,23,37,38,39]. For instance, FSRCNN [22] utilized deconvolution layers to achieve faster processing speed compared to SRCNN.…”

Section: B Lightweight Super-resolution Modelsmentioning

confidence: 99%

TAKDSR: Teacher Assistant Knowledge Distillation Framework for Graphics Image Super-Resolution

Yoon,

Lee,

Song

2023

IEEE Access

View full text Add to dashboard Cite

This paper presents a framework for effectively applying knowledge distillation (KD) to superresolution (SR) tasks for computer graphics (CG) images. Specifically, we propose TAKDSR, a KD framework for SR using a teacher assistant (TA) network. Recently, the performance of SR models has improved dramatically thanks to the development of deep learning. SR models have evolved into a form that requires a considerable amount of computation and parameters while adopting a complex neural network structure to improve performance. However, it is difficult to utilize conventional high-performance SR models for real-time up-scaling in CG applications requiring high resolution and high frame rate. To solve this, we employ an approach that applies KD to a lightweight SR model. At this time, if the high-resolution (HR) image is used as input for the teacher to show superior performance to the student, a large performance difference occurs between the two due to the excessive performance of the teacher. As a result, the teacher's knowledge has a significantly hard and complex nature, and when transferred to the student, the effect of KD can be rather weakened. Therefore, we adopt a TA network to facilitate the propagation of knowledge between teacher and student. At the same time, the distribution of compact features (CF), which are the decoder input of the teacher, is discretized so that it is compatible with the input distribution of the student, enabling effective KD. Experimental results demonstrate the proposed TAKDSR significantly improves the performance of a given SR model on CG image datasets.

show abstract

Section: B Lightweight Super-resolution Modelsmentioning

confidence: 99%

TAKDSR: Teacher Assistant Knowledge Distillation Framework for Graphics Image Super-Resolution

Yoon,

Lee,

Song

2023

IEEE Access

View full text Add to dashboard Cite

show abstract

“…There are various upsampling methods. Traditional methods such as deconvolution and uppooling tend to generate high computational complexity and introduce more additional information that are not conducive to image reconstruction [23]. In contrast, PixelShuffle is a more effective upsampling method.…”

Section: The Coloring Subnetmentioning

confidence: 99%

Adversarial Image Colorization Method Based on Semantic Optimization and Edge Preservation

Gui

Zhan²,

Gu³

et al. 2022

Electronics

Self Cite

View full text Add to dashboard Cite

As a medium for transmitting visual information, image is a direct reflection of the objective existence of the natural world. Grayscale images lack more visual information than color images. Therefore, it is of great significance to study the colorization of grayscale images. At present, the problems of semantic ambiguity, boundary overflow and lack of color saturation exist in both traditional and deep learning methods. To solve the above problems, an adversarial image colorization method based on semantic optimization and edge preservation is proposed. By improving generative and discriminative networks and designing loss functions, deeper semantic information and sharper edges of images can be learned by our network. Our experiments are carried out on the public datasets Place365 and ImageNet. The experimental results show that the method in this paper can reduce the color anomaly caused by semantic ambiguity, suppress the color blooming in the image boundary area and improve the saturation of the image. Our work achieves competitive results on objective indicators of peak signal-to-noise ratio (PSNR), structural similarity (SSIM) and learned perceptual image patch similarity (LPIPS), with values of 30.903 dB, 0.956 and 0.147 on Place365 and 30.545 dB, 0.946 and 0.150 on ImageNet, which proves that this method can effectively colorize grayscale images.

show abstract

“…One of the most popular topics in vision is multimodal image registration; in particular, infrared and visible image registration is the most studied [3,4]. Since infrared and visible images can provide complementary information, they have wide applications in image fusion [5,6], automatic transform detection [7], and super-resolution reconstruction [8].…”

Section: Introductionmentioning

confidence: 99%

RI-MFM: A Novel Infrared and Visible Image Registration with Rotation Invariance and Multilevel Feature Matching

Zhu

Zhan

et al. 2022

Electronics

Self Cite

View full text Add to dashboard Cite

In the past ten years, multimodal image registration technology has been continuously developed, and a large number of researchers have paid attention to the problem of infrared and visible image registration. Due to the differences in grayscale distribution, resolution and viewpoint between two images, most of the existing infrared and visible image registration methods are still insufficient in accuracy. To solve such problems, we propose a new robust and accurate infrared and visible image registration method. For the purpose of generating more robust feature descriptors, we propose to generate feature descriptors using a concentric-circle-based feature-description algorithm. The method enhances the description of the main direction of feature points by introducing centroids, and, at the same time, uses concentric circles to ensure the rotation invariance of feature descriptors. To match feature points quickly and accurately, we propose a multi-level feature-matching algorithm using improved offset consistency for matching feature points. We redesigned the matching algorithm based on the offset consistency principle. The comparison experiments with several other state-of-the-art registration methods in CVC and homemade datasets show that our proposed method has significant advantages in both feature-point localization accuracy and correct matching rate.

show abstract

Lightweight Dual-Stream Residual Network for Single Image Super-Resolution

Cited by 10 publications

References 48 publications

TAKDSR: Teacher Assistant Knowledge Distillation Framework for Graphics Image Super-Resolution

TAKDSR: Teacher Assistant Knowledge Distillation Framework for Graphics Image Super-Resolution

Adversarial Image Colorization Method Based on Semantic Optimization and Edge Preservation

RI-MFM: A Novel Infrared and Visible Image Registration with Rotation Invariance and Multilevel Feature Matching

Contact Info

Product

Resources

About