Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks108 from public data
- Image demosaicing: a systematic survey424
A systematic survey of over seventy published works in this field since 1999 is provided, which suggests most existing works belong to the class of sequential demosaicing - i.e., luminance channel is interpolated first and then chrominance channels are reconstructed based on recovered luminance information.
- 335
- Aspect Term Extraction with History Attention and Selective Transformation290
Experimental results over four benchmark datasets clearly demonstrate that the new framework for tackling ATE can outperform all state-of-the-art methods.
- 123
- 91
- LIQA: Lifelong Blind Image Quality Assessment87
This work proposes a novel Lifelong blind Image Quality Assessment (LIQA) approach, targeting to achieve the lifelong learning of BIQA, and adopts the Split-and-Merge distillation strategy to train a single-head network that makes task-agnostic predictions.
- Learned Block-Based Hybrid Image Compression74
This paper introduces explicit intra prediction into a learned image compression framework to utilize the relation among adjacent blocks and proposes a contextual prediction module (CPM) to better capture long-range correlations by utilizing the strip pooling to extract the most relevant information in neighboring latent space, thus achieving effective information prediction.
- 59
- 58
- Task-Driven Semantic Coding via Reinforcement Learning54
This work designs semantic maps for different tasks to extract the pixelwise semantic fidelity for videos/images and implements task-driven semantic coding by implementing semantic bit allocation based on reinforcement learning (RL).
- Learning Omni-Frequency Region-adaptive Representations for Real Image Super-Resolution53
This paper designs a Frequency Decomposition module to separate different frequency components to comprehensively compensate the information lost for real LR image and designs a Region-adaptive Frequency Aggregation module by leveraging dynamic convolution and spatial attention to adaptively restore frequency components for different regions.
- Learning Disentangled Feature Representation for Hybrid-Distorted Image Restoration50
This work introduces the concept of Disentangled Feature Learning to achieve the feature-level divide-and-conquer of hybrid distortions and proposes the feature disentanglement module (FDM) to distribute feature representations of different distortions into different channels by revising gain-control-based normalization.
- Convolutional‐capsule network for gastrointestinal endoscopy image classification45
The core of the method is a lesion‐aware CNN feature extraction module which can encode sufficiently detailed information of lesions in midlevel CNN features and in turn enable the subsequent capsule classification network to effectively learn deformation‐invariant relationships between image entities.
- NTIRE 2021 Challenge on Quality Enhancement of Compressed Video: Methods and Results43
This paper reviews the first NTIRE challenge on quality enhancement of compressed video, with a focus on the proposed methods and results, and gauge the state of the art of video quality enhancement.
- AIM 2020 Challenge on Real Image Super-Resolution: Methods and Results43
This paper introduces the real image Super-Resolution challenge that was part of the Advances in Image Manipulation (AIM) workshop, held in conjunction with ECCV 2020, and gauges the state-of-the-art approaches for real image SR in terms of PSNR and SSIM.
- 40
- PromptCIR: Blind Compressed Image Restoration with Prompt Learning39
A prompt-learning-based compressed image restoration network, dubbed PromptCIR, which can effectively restore images from various compress levels, and exploits prompts to encode compression information implicitly, where prompts directly interact with soft weights generated from image features, thus providing dynamic content-aware and distortion-aware guidance for the restoration process.
- Learning Distortion Invariant Representation for Image Restoration from a Causality Perspective33
This paper is the first to propose a novel training strategy for image restoration from the causality perspective, to improve the generalization ability of DNNs for unknown degradations.
- FAN: Frequency Aggregation Network for Real Image Super-Resolution32
This work proposes FAN, a frequency aggregation network, to address the real-world image super-resolu-tion problem by extracting different frequencies of the LR image and passing them to a channel attention-grouped residual dense network (CA-GRDB) individually to output corresponding feature maps.
- LIRA: Lifelong Image Restoration from Unknown Blended Distortions27
Experimental results show that the proposed approach can not only achieve state-of-the-art performance on blended distortions removal tasks in both PSNR/SSIM metrics, but also maintain old expertise while learning new restoration tasks.
- StyleAM: Perception-Oriented Unsupervised Domain Adaption for No-Reference Image Quality Assessment26
This work proposes an effective perception-oriented unsupervised domain adaptation method StyleAM (Style Alignment and Mixup) for NR-IQA, which transfers sufficient knowledge from label-rich source domain data to label-free target domain images.
- 26
- Multi-scale Grouped Dense Network for VVC Intra Coding24
The multi-scale grouped dense network (MSGDN) is designed to further reduce the compression artifacts by combining the multi- scale and grouped dense block, which are integrated as the post-process network of VVC intra coding.
- 21
- 21
- Method for rectifying image deviation based on perspective transformation19
Experimental results show that the algorithm can realize image rectification, which is simple with fast speed and high precision, and the proposed method is helpful for the further recognition.
- 17
- Local Patch AutoAugment With Multi-Agent Collaboration17
This paper proposes a more fine-grained automated DA approach, dubbed Patch AutoAugment, to divide an image into a grid of patches and search for the joint optimal augmentation policies for the patches as a multi-agent reinforcement learning (MARL) problem.
- 15
- QoS-Aware Publish-Subscribe Service for Real-Time Data Acquisition15
A novel QoS-aware Real-Time Publish-Subscribe (QRTPS) service compatible to DDS for distributed real-time data acquisition is proposed and implemented on Agilor by using objects and RECA rules in AgILor.
- 15
- 13
- FreqAlign: Excavating Perception-Oriented Transferability for Blind Image Quality Assessment From a Frequency Perspective12
This work proposes an effective alignment strategy, i.e., Frequency Alignment (dubbed FreqAlign), to excavate the perception-oriented transferability of BIQA in the frequency space by performing feature frequency decomposition and selecting the frequency components that contained the most transferable perception knowledge for alignment.
- 12
- 11
- 11
- 11
- 11
- 10
- 10
- Single image super resolution via wavelet transform fusion and SRFeat network10
A novel single image super resolution method based on SRFeat network and wavelet fusion is proposed that can not only reduce the probability of image distortion, but recover the global information of the reconstructed image and remove the noise brought by the reconstruction process.
- 10
- 9
- Semantic-Aware Message Broadcasting for Efficient Unsupervised Domain Adaptation8
A novel method, Semantic-aware Message Broadcasting (SAMB), which enables more informative and flexible feature alignment for unsupervised domain adaptation (UDA) and systematically study the effects of adversarial-based feature alignment (ADA) and pseudo-label based self-training (PST) on UDA.
- 7
- A fully open structure-guided RNA foundation model for robust structural and functional inference6
The proposed structRFM model is a structure-guided RNA foundation model that is pre-trained on millions of RNA sequences and secondary structures data by integrating base pairing interactions into masked language modeling through a novel pair matching operation, and sets new benchmarks for secondary structure prediction.
- Application of Improved Harris Algorithm in Sub-Pixel Feature Point Extraction6
According to mathematics principles, threshold calculation method for the algorithm has been improved and the influence of k selected by users on the stability of the algorithm is eliminated and this method is effective in improving the precision and accuracy of feature point positioning.
- 5
- 5
- A Method of Multi-Attribute Synthetic Evaluation Based on Cloud Model5
This paper maps the indefinite comment as different cloud droplets with fine variations, and maps the variation of each factor’s weight as correctable mathematic expectation of the cloud, and thus puts forward a comprehensive evaluation model based on cloud model.
- A robust eye localization algorithm for face recognition5
An algorithm is proposed for eye localization that can accurately locate the eyes, and it is robust to the variations of face poses, illuminations, expressions, and accessories.
- 4
- Directional lifting-based wavelet transform for multiple description image coding4
A new architecture for MD image coding is proposed, which is based on directional wavelet transform that removes redundancy among spatial and wavelet domain and reconstructs one description with full resolution.
- 3
- The Study of PET Image Segmentation Using Clustering Algorithm3
Two kinds of clustering algorithms, K-means algorithm and the improved algorithm based on ISODATA, were validated and compared, with the simulated PET data and clinical dynamic PET images, to separate the region of interest.
- 3
- Mining Frequent Patterns in Print Logs with Semantically Alternative Labels3
This paper proposes a novel pattern mining problem, namely mining frequent patterns with semantically alternative labels, and proposes an efficient algorithm named PaSAL Frequent Patterns with Semantically Alternative Labels Mining Algorithm for this problem, which is evaluated on a real print log data.
- 3
- Constructing near-tight wavelet frames by neural networks3
It is shown that a simple linear combination of dilates and translates of (sigma) generates a near tight wavelet frame for L2(R), which is then used in constructing approximation to multivariate functions by neural networks with one hidden layer.
- A close look at few-shot real image super-resolution from the distortion relation perspective2
This work is the first to have a close look at the under-explored direction for RealSR, i.e., few-shot real-world image super-resolution, and proposes Distortion Relation guided Transfer Learning (DRTL) for the few- shot RealSR by transferring the rich restoration knowledge from auxiliary distortions to the target RealSR under the guidance of distortion relation.
- 2
- 2
- Mitigating Object Hallucinations in Large Vision-Language Models through Visual Contrastive Decoding2
- Neighborhood Rough Fuzzy Penetration Control Method with Variable Precision Based on GMAW2
A variable precision neighborhood rough-fuzzy control model is proposed to enhance the efficiency and adaptability of rough set theory for information processing in the welding process and has excellent stability and effectiveness.
- 2
- 2
- 2
- 2
- 2
- 2
- Deep learning-based 6-DoF visual relocalization assisted simultaneous localization and mapping (SLAM)1
This study proposes a complete DL-based coarse-to-fine 6-DoF long-term visual relocalization method to assist vSLAM, which demonstrates enhanced robustness and generalizability and can eliminate cumulative errors in pose estimation under challenging environments.
- Variable Gap MAG Welding Penetration Control Using Rough Set Knowledge Acquisition1
This article presents a scheme of variable gap welding penetration control based on rough-fuzzy modeling, utilizing the basic concepts of variable precision rough sets and conditional information entropy to address relevant algorithms for attribute reduction and rule extraction in decision information systems.
- 1
- Fault Diagnosis of Multi-Information Fusion in Train Intelligent Control System Based on Fuzzy Neutral Network and Evidential Theory1
Result of example indicates that this method of fault diagnosis of multi-information fusion in train intelligent control system based on fuzzy neural network and D-S evidential theory can effectively improve the probability and reduce the uncertainty of diagnosis.
- Comparing both relevance and robustness in selection of web ranking functions1
It is argued that the ranking robustness can be effectively measured by taking into account the ranking score distribution across Web pages, and NDCG is improved with two new metrics and shows their superiority in terms of stability to ranking score turbulence and stability in function selection.
- Bayesian image decomposition applied to relaxographic imaging1
The method, Bayesian Image Decomposition, recovers individual tissue type magnetization recovery curves and their corresponding tissue‐specific relaxographic images (i.e. segmented images) from a series of inversion recovery images.
- 1
- –
- Joint Degradation-Aware Arbitrary-Scale Super-Resolution for Variable-Rate Extreme Image Compression–
ASSR-EIC is proposed, a novel image compression framework that leverages arbitrary-scale super-resolution (ASSR) to support variable-rate extreme image compression (EIC) and introduces a dual semantic-enhanced design that offers holistic guidance for rate adaptation and adaptive rate-dependent reconstruction.
- –
- –
- –
- –
- –
- –
- –
- Biexponential fitting to filter CSF-rich voxels and denoise diffusion-weighted image data: improving MADI quantification–
This work improves MADI accuracy by robust filtering of CSF-rich voxels and denoising diffusion-weighted images using biexponential fitting, improving MADI quantitative maps.
- –
- –
- –
- –
- Optimizing Prostate DWI Acquisitions for non-Gaussian models using an Activity MRI [aMRI] library–
Using the best-matched curves from a DWI library created with Monte Carlo random walk simulations of contracted Voronoi-cell ensembles, this work shows that, for prostate DWI, a maximum b-value near 2000 s/mm2 is optimal and the number of b-values is not as crucial as long as it is larger than the numberof model parameters.
- MRCP at 1.5 T: comparing image quality and acquisition time between respiratory-triggered and breath-hold SPACE sequences–
The results suggested that modified RT-SPACE-MRCP sequence can significantly reduced acquisition time and had better image quality.
- –
- –
- –
- Image tracking technology for dynamic angle measurement of machine vision–
High accuracy and good real-time performance of this method in processing the target image is demonstrated by calculating and comparing the results of each image processing algorithm, which satisfies the measured requirement of space dynamic angle.
- –
- –
- Design and FPGA implementation of real-time automatic image enhancement algorithm–
The algorithm is based on the design and implementation of FPGA hardware circuit less cost on hardware, high real-time performance, the good processing performance in different sceneries, and would have wide prospect on imaging processing field.
- Study on the application of four- dimensional spatio- temporal image correlation in acquisition of nine fetal echocardiographic views–
Multiple-view pattern is a recommendable method, the quality of images is satisfactory, and the display rates in fetuses more than 33 gestational weeks are relatively low; although the technology has a certain disadvantages,fetal cardiac examination with STIC was also a beneficial supplemental method for conventional fetal echocardiography.
- An efficient coding method of medical image data based on wavelet transforms–
Aiming at the different local feature information of the image signal, the coding and compression of it are realized by two-dimensional wavelet transform in wavelet method so more information can be expressed as much fewer codes as possible.
- WSN Image Watermarking Algorithm Based on Multi-Wavelet–
In this paper, the wavelet transform coefficients were analyzed by the noise visibility function and bits were embedded into the coefficients which have more ability to the tolerance of the noise according to the designated threshold.
- Multiple Description Image Coding Based on Directional Wavelet Transform and RKF Interpolation–
The proposed architecture is based on directional wavelet transform that removes redundancy among wavelet domain and a new interpolation method is used to reconstruct the image when one of the description is lost.
- Watermark Image Restoration Method Based on Block Hopfield Network–
Hopfield network is introduced for the restoration of extracted watermark image which may be blurred due to the signal transfer or various signal processing operations and a novel codebook method is designed to reduce the storage space and to increase the security.
- Vision-based augmented reality computer assisted surgery navigation system–
A vision-based Augmented Reality computer assisted surgery navigation system that applies theAugmented Reality technique to surgery Navigation system, so the surgeon's vision of the real world is enhanced.
- Incremental query processing in information fusion systems–
Aiming to remove the query building bottleneck, a novel incremental query method is proposed, which can eliminate the accumulated complexity in query building as well as in query execution.
- Design JXTA Based Super-peer Search Method–
This paper compares the advantage and deficiency of these three kinds of search methods, and based on JX-TA platform design the super-peer search method.
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-12. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.