/pdf/srdiff-single-image-super-resolution-with-diffusion-i00abuwt.pdf

SRDiff: Single image super-resolution with diffusion probabilistic models

Low-light image enhancement aims to recover normal-light images from the images captured under very dim environments. Existing methods cannot well handle the noise, color bias and over-exposure problem, and fail to ensure visual quality when lacking paired training data. To address these problems, we propose a novel unsupervised low-light image enhancement network named LE-GAN, which is based on generative adversarial networks and is trained with unpaired low/normal-light images. Specifically, we design an illumination-aware attention module that enhances the feature extraction of the network to address the problems of noise and color bias, as well as improve the visual quality. We further propose a novel identity invariant loss to address the over-exposure problem to make the network learn to enhance low-light images adaptively. Extensive experiments show that the proposed method can achieve promising results. Furthermore, we collect a large-scale low-light dataset named Paired Normal/Lowlight Images (PNLI). It consists of 2,000 pairs of low/normal-light images captured in various real-world scenes, which can provide the research community with a high-quality dataset to advance the development of this field. 

LE-GAN: Unsupervised low-light image enhancement network using attention module and identity invariant loss

This article propose a parameter estimation method for of a damped sinusoidal signal by the frequency domain interpolation. The three peaks of the discrete Fourier transform (DFT) result are used in the interpolation algorithm to estimate the damping factor and signal frequency. The proposed method can achieve a high accuracy because the influence of spectrum leakage from negative frequency are completely removed during the frequency domain interpolation. The comparison of different algorithms are discussed, which confirm the effectiveness of the proposed method.

Estimation of Damping Factor and Signal Frequency for Damped Sinusoidal Signal by Three Points Interpolated DFT

Improving recommender system via knowledge graph based exploring user preference

Convolutional neural networks (CNNs) have obtained remarkable performance via deep architectures. However, these CNNs often achieve poor robustness for image super-resolution (SR) under complex scenes. In this article, we present a heterogeneous group SR CNN (HGSRCNN) via leveraging structure information of different types to obtain a high-quality image. Specifically, each heterogeneous group block (HGB) of HGSRCNN uses a heterogeneous architecture containing a symmetric group convolutional block and a complementary convolutional block in a parallel way to enhance the internal and external relations of different channels for facilitating richer low-frequency structure information of different types. To prevent the appearance of obtained redundant features, a refinement block (RB) with signal enhancements in a serial way is designed to filter useless information. To prevent the loss of original information, a multilevel enhancement mechanism guides a CNN to achieve a symmetric architecture for promoting expressive ability of HGSRCNN. Besides, a parallel upsampling mechanism is developed to train a blind SR model. Extensive experiments illustrate that the proposed HGSRCNN has obtained excellent SR performance in terms of both quantitative and qualitative analysis. Codes can be accessed at https://github.com/hellloxiaotian/HGSRCNN.

A heterogeneous group CNN for image super-resolution

Sequential recommendations aim to predict the users next behaviors items based on their successive historical behaviors sequence. It has been widely applied in lots of online services. However, current sequential recommendations use the adjacent behaviors to capture the features of the sequence, ignoring the features among nonadjacent sequential items and the summarized features of the sequence. To address the above problems, in this paper, we propose an annular-graph attention based sequential recommendation (AGSR) model by exploring users long-term and short-term preferences for the personalized sequential recommendation. For users short-term preferences, AGSR builds an annular-graph on the sequence of user behavior. Then, AGSR proposes an annular-graph attention applying on the sub annular-graph to explore local features and applying annular-graph attention on entire annular-graph to explore the global features and the skip features. For users long-term preferences, the latent factor model are introduced in AGSR. The experimental results on two public datasets show that our model outperforms the state-of-the-art methods.

Annular-Graph Attention Model for Personalized Sequential Recommendation

Single image super-resolution is very important as a low-level computer vision task. With the development of deep convolution neural networks (CNNs), recent approaches with CNNs have outperformed existing traditional methods in the single image super-resolution (SISR) field. However, these methods may suffer from weaker representational power and overly-smoothing textures. To handle these problems, we propose a Kernel-Attended Residual Network (KARN). Our KARN possesses the optimal performance for feature fusion and feature representation. Specifically, we present a multi-channel fusion block (MCFB) to restore plentiful textual feature information, and a kernel-attended block (KAB) to improve the representation power of our network with multiple kernels. Besides, we present a space-feature re-calibration block (SFRB) to integrate the calibration into features in the spatial aspect. Owing to the advanced information that we extract, KARN achieves a more notable performance than state-of-the-art methods by evaluating the performance of results based on benchmark datasets.

Kernel-attended residual network for single image super-resolution

Image restoration (IR) aims to recover image quality from various degradations. Existing convolutional neural networks (CNN) based IR methods try to improve performance by enlarging the model receptive field with the sacrifice of fine spatial details and extra artifacts. This paper proposes a Deformable and Attentive Network (DANet) to address these problems. In DANet, we propose two novel blocks: Attentive DEformable-convolution Block (ADEB) and Attentive Recurrent Offset Block (AROB). In ADEB, deformable convolution is collaborated with various attention modules to generate more adaptive receptive fields. AROB transfers more attentive texture information among different scales during the encoding/decoding process for ADEB. To further refine DANet, we propose a knowledge distillation scheme to train a light-weighted DANet (DANet-S) with limited performance loss. Extensive experiments on several image benchmark datasets demonstrate that our method achieves state-of-the-art (SOTA) results for various IR tasks, including image denoising, JPEG artifacts removal, and real-world super resolution.

Learning Deformable and Attentive Network for image restoration

Although many frequency estimation methods are available, few are designed for high-quality speech and audio processing systems, which typically use the modified discrete cosine transform (MDCT) as their analysis filter bank. In this letter, we propose a low complexity frequency estimator that is suitable for MDCT-based systems and that operates in the odd-DFT domain. Taking a complex exponential in noise as the input and deriving the analytical expression of its odd-DFT coefficient, we obtain an interpolated odd-DFT-based frequency estimator. Experiments show that the proposed estimator outperforms all other reported odd-DFT/MDCT domain estimators and has precision that is similar to that of representative DFT domain frequency estimators. The overhead for incorporating this estimator into a speech and audio processing system is small due to the simple odd-DFT to MDCT conversion. The corresponding magnitude and phase estimators are also proposed in this letter.

Yujie Dun

Papers

Annular-Graph Attention Model for Personalized Sequential Recommendation

Kernel-attended residual network for single image super-resolution

Learning Deformable and Attentive Network for image restoration

A Fine-Resolution Frequency Estimator in the Odd-DFT Domain

Image super-resolution based on residually dense distilled attention network