Image Super-Resolution via Adaptive Regularization and Sparse Representation

Road Network-Guided Fine-Grained Urban Traffic Flow Inference

Accurate inference of fine-grained traffic flow from coarse-grained one is an emerging yet crucial problem, which can help greatly reduce the number of the required traffic monitoring sensors for cost savings. In this work, we note that traffic flow has a high correlation with road network, which was either completely ignored or simply treated as an external factor in previous works. To facilitate this problem, we propose a novel road-aware traffic flow magnifier (RATFM) that explicitly exploits the prior knowledge of road networks to fully learn the road-aware spatial distribution of fine-grained traffic flow. Specifically, a multidirectional 1-D convolutional layer is first introduced to extract the semantic feature of the road network. Subsequently, we incorporate the road network feature and coarse-grained flow feature to regularize the short-range spatial distribution modeling of road-relative traffic flow. Furthermore, we take the road network feature as a query to capture the long-range spatial distribution of traffic flow with a transformer architecture. Benefiting from the road-aware inference mechanism, our method can generate high-quality fine-grained traffic flow maps. Extensive experiments on three real-world datasets show that the proposed RATFM outperforms state-of-the-art models under various scenarios. Our code and datasets are released at https://github.com/luimoli/RATFM.

Regularization by denoising: Bayesian model and Langevin-within-split Gibbs sampling

This paper introduces a Bayesian framework for image inversion by deriving a probabilistic counterpart to the regularization-by-denoising (RED) paradigm. It additionally implements a Monte Carlo algorithm specifically tailored for sampling from the resulting posterior distribution, based on an asymptotically exact data augmentation (AXDA). The proposed algorithm is an approximate instance of split Gibbs sampling (SGS) which embeds one Langevin Monte Carlo step. The proposed method is applied to common imaging tasks such as deblurring, inpainting and super-resolution, demonstrating its efficacy through extensive numerical experiments. These contributions advance Bayesian inference in imaging by leveraging data-driven regularization strategies within a probabilistic framework.

Are Graph Convolutional Networks With Random Weights Feasible?

Graph Convolutional Networks (GCNs), as a prominent example of graph neural networks, are receiving extensive attention for their powerful capability in learning node representations on graphs. There are various extensions, either in sampling and/or node feature aggregation, to further improve GCNs' performance, scalability and applicability in various domains. Still, there is room for further improvements on learning efficiency because performing batch gradient descent using the full dataset for every training iteration, as unavoidable for training (vanilla) GCNs, is not a viable option for large graphs. The good potential of random features in speeding up the training phase in large-scale problems motivates us to consider carefully whether GCNs with random weights are feasible. To investigate theoretically and empirically this issue, we propose a novel model termed Graph Convolutional Networks with Random Weights (GCN-RW) by revising the convolutional layer with random filters and simultaneously adjusting the learning objective with regularized least squares loss. Theoretical analyses on the model's approximation upper bound, structure complexity, stability and generalization, are provided with rigorous mathematical proofs. The effectiveness and efficiency of GCN-RW are verified on semi-supervised node classification task with several benchmark datasets. Experimental results demonstrate that, in comparison with some state-of-the-art approaches, GCN-RW can achieve better or matched accuracies with less training time cost.

Noise Robust Face Hallucination Based on Smooth Correntropy Representation

Face hallucination technologies have been widely developed during the past decades, among which the sparse manifold learning (SML)-based approaches have become the popular ones and achieved promising performance. However, these SML methods always failed in handling noisy images due to the least-square regression (LSR) they used for error approximation. To this end, we propose, in this article, a smooth correntropy representation (SCR) model for noisy face hallucination. In SCR, the correntropy regularization and smooth constraint are combined into one unified framework to improve the resolution of noisy face images. Specifically, we introduce the correntropy induced metric (CIM) rather than the LSR to regularize the encoding errors, which admits the proposed method robust to noise with uncertain distributions. Besides, the fused LASSO penalty is added into the feature space to ensure similar training samples holding similar representation coefficients. This encourages the SCR not only robust to noise but also can well exploit the inherent typological structure of patch manifold, resulting in more accurate representations in noise environment. Comparison experiments against several state-of-the-art methods demonstrate the superiority of SCR in super-resolving noisy low-resolution (LR) face images.

Deep Blind Hyperspectral Image Super-Resolution

The production of a high spatial resolution (HR) hyperspectral image (HSI) through the fusion of a low spatial resolution (LR) HSI with an HR multispectral image (MSI) has underpinned much of the recent progress in HSI super-resolution. The premise of these signs of progress is that both the degeneration from the HR HSI to LR HSI in the spatial domain and the degeneration from the HR HSI to HR MSI in the spectral domain are assumed to be known in advance. However, such a premise is difficult to achieve in practice. To address this problem, we propose to incorporate degeneration estimation into HSI super-resolution and present an unsupervised deep framework for "blind" HSIs super-resolution where the degenerations in both domains are unknown. In this framework, we model the latent HR HSI and the unknown degenerations with deep network structures to regularize them instead of using handcrafted (or shallow) priors. Specifically, we generate the latent HR HSI with an image-specific generator network and structure the degenerations in spatial and spectral domains through a convolution layer and a fully connected layer, respectively. By doing this, the proposed framework can be formulated as an end-to-end deep network learning problem, which is purely supervised by those two input images (i.e., LR HSI and HR MSI) and can be effectively solved by the backpropagation algorithm. Experiments on both natural scene and remote sensing HSI data sets show the superior performance of the proposed method in coping with unknown degeneration either in the spatial domain, spectral domain, or even both of them.

Survey on Multi-Output Learning

The aim of multi-output learning is to simultaneously predict multiple outputs given an input. It is an important learning problem for decision-making since making decisions in the real world often involves multiple complex factors and criteria. In recent times, an increasing number of research studies have focused on ways to predict multiple outputs at once. Such efforts have transpired in different forms according to the particular multi-output learning problem under study. Classic cases of multi-output learning include multi-label learning, multi-dimensional learning, multi-target regression, and others. From our survey of the topic, we were struck by a lack in studies that generalize the different forms of multi-output learning into a common framework. This article fills that gap with a comprehensive review and analysis of the multi-output learning paradigm. In particular, we characterize the four Vs of multi-output learning, i.e., volume, velocity, variety, and veracity, and the ways in which the four Vs both benefit and bring challenges to multi-output learning by taking inspiration from big data. We analyze the life cycle of output labeling, present the main mathematical definitions of multi-output learning, and examine the field's key challenges and corresponding solutions as found in the literature. Several model evaluation metrics and popular data repositories are also discussed. Last but not least, we highlight some emerging challenges with multi-output learning from the perspective of the four Vs as potential research directions worthy of further studies.

A novel diagnostic information based framework for super-resolution of retinal fundus images

Advancements in tele-medicine have led to the development of portable and cheap hand-held retinal imaging devices. However, the images obtained from these devices have low resolution (LR) and poor quality that may not be suitable for retinal disease diagnosis. Therefore, this paper proposes a novel framework for the super-resolution (SR) of the LR fundus images. The method takes into consideration the diagnostic information in the fundus images during the SR process. In this work, SR is performed on the zone of interest of the fundus images. Clinical information of the selected zone is captured using the Shannon entropy, the contrast sensitivity index (CSI), the multi-resolution (MR) intra-band energy and the MR inter-band eigen features. The support vector machine (SVM) classifier is used to decide the clinical significance of the zone. Highly accurate learning based SR method or the bicubic interpolation is applied to the selected zone based on the classification output. The method is tested on the Standard Diabetic Retinopathy Database Calibration level 1 (DIARETDB1) and the Digital Retinal Images for Vessel Extraction (DRIVE) databases. Classification accuracy of 85.22% and 85.77% is achieved for the DIARETDB1 and DRIVE databases respectively. The SR performance of the algorithm is quantified in terms of peak signal to noise ratio (PSNR), structural similarity index measure (SSIM) and computational time. The proposed diagnostic information based SR achieves computational time efficiency without compromising with the high resolution (HR) reconstruction accuracy of the fundus image zones.

Robust DLPP With Nongreedy $\ell _1$ -Norm Minimization and Maximization

Recently, discriminant locality preserving projection based on L1-norm (DLPP-L1) was developed for robust subspace learning and image classification. It obtains projection vectors by greedy strategy, i.e., all projection vectors are optimized individually through maximizing the objective function. Thus, the obtained solution does not necessarily best optimize the corresponding trace ratio optimization algorithm, which is the essential objective function for general dimensionality reduction. It results in insufficient recognition accuracy. To tackle this problem, we propose a nongreedy algorithm to solve the trace ratio formula of DLPP-L1, and analyze its convergence. Experimental results on three databases illustrate the effectiveness of our proposed algorithm.