Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks76 from public data
- Gated Fusion Network for Single Image Dehazing953
An efficient algorithm to directly restore a clear image from a hazy input using an end-to-end trainable neural network that consists of an encoder and a decoder is proposed.
- Unsupervised Image Super-Resolution Using Cycle-in-Cycle Generative Adversarial Networks619
This work proposes a Cycle-in-Cycle network structure with generative adversarial networks (GAN) as the basic component to tackle the single image super-resolution problem in a more general case that the low-/high-resolution pairs and the down-sampling process are unavailable.
- CREST: Convolutional Residual Learning for Visual Tracking553
This paper proposes the CREST algorithm to reformulate DCFs as a one-layer convolutional neural network, and applies residual learning to take appearance changes into account to reduce model degradation during online update.
- 409
- DNGaussian: Optimizing Sparse-View 3D Gaussian Radiance Fields with Global-Local Depth Normalization383
DNGaussian is introduced, a depth-regularized framework based on 3D Gaussian radiance fields, offering real-time and high-quality few-shot novel view synthesis at low costs, and a Hard and Soft Depth Regularization to restore accurate scene geometry under coarse monocular depth supervision while maintaining a fine-grained color appearance.
- Spatio-Temporal Filter Adaptive Network for Video Deblurring237
The proposed Spatio-Temporal Filter Adaptive Network (STFAN) takes both blurry and restored images of the previous frame as well as blurry image of the current frame as input, and dynamically generates the spatially adaptive filters for the alignment and deblurring.
- Learning Dual Convolutional Neural Networks for Low-Level Vision221
The proposed DualCNN consists of two parallel branches, which respectively recovers the structures and details in an end-to-end manner, and can generate the target signals according to the formation model for each particular application.
- Learning Fully Convolutional Networks for Iterative Non-blind Deconvolution200
The proposed fully convolutional network is able to learn an adaptive image prior, which keeps both local and global information, and performs favorably against state-of-the-art algorithms in terms of quality and speed.
- Physics-Based Generative Adversarial Models for Image Restoration and Beyond188
An algorithm to directly solve numerous image restoration problems (e.g., image deblurring, image dehazing, and image deraining) by generative models with adversarial learning within the GAN framework.
- RestoreFormer: High-Quality Blind Face Restoration from Undegraded Key-Value Pairs177
This work proposes a method, RestoreFormer, which explores fully-spatial attentions to model contextual information and surpasses existing works that use local operators and outperforms advanced state-of-the-art methods on one synthetic dataset and three real-world datasets.
- 120
- DAVANet: Stereo Deblurring With View Aggregation108
This work proposes a novel stereo image deblurring network with Depth Awareness and View Aggregation, named DAVANet, which outperforms state-of-the-art methods in terms of accuracy, speed, and model size.
- Learning to Hallucinate Face Images via Component Generation and Enhancement107
Facial components are generated to approximate ground truth global appearance in the first stage and enhance them through recovering details in the second stage to perform favorably against state-of-the-art methods.
- Grassmann Pooling as Compact Homogeneous Bilinear Pooling for Fine-Grained Visual Classification102
An alternative pooling method which transforms the CNN feature matrix to an orthonormal matrix consists of its principal singular vectors, which enables a very compact feature and classifier representation on a variety of fine-grained image classification datasets.
- Deep Blind Video Super-resolution98
An effective blind video SR algorithm based on deep convolutional neural networks (CNNs) and an effective sharp feature exploration method which first extracts sharp features from restored intermediate latent frames and then uses a transformation operation based on the extracted sharp features and warped features from LR inputs to generate better features for HR video restoration.
- Deep Non-Blind Deconvolution via Generalized Low-Rank Approximation96
The authors' analysis shows that the estimated decomposed matrices contain the most essential information of the input kernel, which ensures the proposed network to handle various blurs in a unified framework and generate high-quality deblurring results.
- AIM 2020 Challenge on Learned Image Signal Processing Pipeline90
This paper reviews the second AIM learned ISP challenge and provides the description of the proposed solutions and results, defining the state-of-the-art for practical image signal processing pipeline modeling.
- 70
- EfficientFCN: Holistically-Guided Decoding for Semantic Segmentation61
The EfficientFCN is proposed, whose backbone is a common ImageNet pre-trained network without any dilated convolution, and achieves comparable or even better performance than state-of-the-art methods with only 1/3 of the computational cost.
- Joint Face Hallucination and Deblurring via Structure Generation and Detail Enhancement60
This paper proposes a facial component guided deep Convolutional Neural Network to restore a coarse face image, which is denoted as the base image where the facial component is automatically generated from the input face image.
- Dual Convolutional Neural Networks for Low-Level Vision59
Experimental results show that the DualCNN can be effectively applied to numerous low-level vision tasks with favorable performance against the state-of-the-art methods that have been specially designed for each individual task.
- RestoreFormer++: Towards Real-World Blind Face Restoration From Undegraded Key-Value Pairs56
This work proposes RestoreFormer++, which on the one hand introduces fully-spatial attention mechanisms to model the contextual information and the interplay with the priors, and on the other hand explores an extending degrading model to help generate more realistic degraded face images to alleviate the synthetic-to-real-world gap.
- 48
- Diffusion-based Blind Text Image Super-Resolution43
Extensive experiments on synthetic and real-world datasets demonstrate that the Diffusion-based Blind Text Image Super-Resolution (DiffTSR) can restore text images with more accurate text structures as well as more realistic appearances simultaneously.
- 41
- Deep Richardson–Lucy Deconvolution for Low-Light Image Deblurring37
A data-driven approach to model the saturated pixels by a learned latent map is developed and performs favorably against state-of-the-art algorithms both quantitatively and qualitatively on synthetic and real-world images.
- 34
- OID: Outlier Identifying and Discarding in Blind Image Deblurring31
This paper develops a simple yet effective Outlier Identifying and Discard-ing (OID) method, which alleviates limitations in existing Maximum A Posteriori (MAP)-based deblurring models when significant outliers are presented.
- 28
- Learning to Deblur Face Images via Sketch Synthesis26
An effective face deblurring algorithm based on deep convolutional neural networks (CNNs) that is able to deblur face images with favorable performance against state-of-the-art methods.
- 25
- 24
- InsTaG: Learning Personalized 3D Talking Head from Few-Second Video23
InsTaG is introduced, a 3D talking head synthesis framework that allows a fast learning of realistic personalized 3D talking head from few training data and achieves high-quality and fast adaptation while preserving high-level personalization and efficiency.
- VersaT2I: Improving Text-to-Image Models with Versatile Reward23
VersaT2I is presented, a versatile training framework that can boost the performance with multiple rewards of any T2I model using the Low-Rank Adaptation (LoRA), and introduces a gating function to combine multiple quality aspects, which can avoid conflicts between different quality aspects.
- Toward visual quality enhancement of dehazing effect with improved Cycle-GAN22
A Cycle Generative Adversarial Network based on the Homology Isomerism Discriminator-assisted Detail Generation Module (DGM) is proposed to achieve image dehazing under unsupervised training and has attractive results in both visual experience and quantitative metrics under various datasets.
- 18
- 2D–3D cascade network for glioma segmentation in multisequence MRI images using multiscale information18
Results indicate that the proposed 2D-3D cascade network with multiscale information module may be a useful tool for glioma segmentation.
- 18
- Efficient Deep Image Denoising via Class Specific Convolution17
Quantitative and qualitative evaluations on public datasets demonstrate that the proposed method for image denoising based on pixel-wise classification can reduce the computational costs without sacrificing the performance compared to state-of-the-art algorithms.
- Multi-Stage Degradation Homogenization for Super-Resolution of Face Images With Extreme Degradations15
- 13
- Learning Diverse Tone Styles for Image Retouching12
It is argued that learning in a one-dimensional style space could 1) disentangle the retouching styles from the image content, 2) lead to a stable style presentation form, and 3) avoid the spatial disharmony effects.
- Super-resolution image reconstruction using fractional-order total variation and adaptive regularization parameters12
The results of experiments indicated that the proposed hybrid SR model with adaptive regularization parameters has a comparative performance compared with state-of-the-art methods and would be potentially more adaptive for the condition of varied blurred kernels.
- RETRACTED ARTICLE: A ML-based economic protection development level using Decision Tree and Ensemble Algorithms10
The proposed ML Algorithm improves decision-making in the economic protection development level field by effectively balancing cost and benefit considerations and will improve accuracy and efficiency to outperform existing approaches, thereby advancing economic protection evaluation and decision- making processes in this domain.
- Multi-view collective tensor decomposition for cross-modal hashing9
This paper proposes a novel multi-view cross-modal hashing method named Multi-view Collective Tensor Decomposition (MCTD) to mitigate the dual heterogeneity at the same time, which can fully exploit the multimodal multi- view feature while simultaneously discovering multiple separated subspaces by leveraging the data categories as supervision information.
- RPN: Reconciled Polynomial Network Towards Unifying PGMs, Kernel SVMs, MLP and KAN8
A novel deep model named Reconciled Polynomial Network (RPN) is introduced for deep function learning and introduces the parameter reconciliation functions to fabricate a small number of parameters into a higher-order parameter matrix to address the ``curse of dimensionality'' problem caused by the data expansions.
- 8
- Analysis and Benchmarking of Extending Blind Face Image Restoration to Videos7
A Temporal Consistency Network (TCN) cooperated with alignment smoothing to reduce jitters and flickers in restored videos is proposed, a flexible component that can be seamlessly plugged into the most advanced face image restoration algorithms, ensuring the quality of image-based restoration is maintained as closely as possible.
- Generic Attention-model Explainability by Weighted Relevance Accumulation6
A weighted relevancy strategy is proposed, which takes the importance of token values into consideration, to reduce distortion when equally accumulating relevance, to reduce distortion when equally accumulating relevance.
- 6
- 5
- 4
- 3
- 2
- 2
- Frontier and Future Directions2
This book introduced the current research works on broad learning and its applications on online social networks and proposed state-of-the-art algorithms to solve these problems, which are introduced in great detail in this book.
- Image kansei feature classification based on color and texture features2
An approach for classifying natural images into emotional categories (happy vs.sad) was proposed and back propagation neural network was employed to map the low level feature vectors to emotional feature space.
- 1
- 1
- 1
- 1
- 1
- –
- –
- CurvLoc: Surface Curvature Prompted Gaussian Splatting for Visual Localization–
This work adopts a 3D Gaussian Splatting representation equipped with surface curvature for the scene, and introduces a novel refinement framework termed CurvLoc, which surpasses existing state-of-the-art methods in accuracy and robustness.
- –
- I 2 TTS: Image-Indicated Immersive Text-to-Speech Synthesis with Spatial Perception–
Embedded Immersive Text-to-speech Synthesis is proposed, an end-to-end multimodal TTS framework that synthesizes high-quality, immersive speech from text and visual scene prompts that produces spatially accurate and natural-sounding speech, achieving superior performance in both subjective and objective evaluations.
- –
- –
- RPN 2: On Interdependence Function Learning Towards Unifying and Advancing CNN, RNN, GNN, and Transformer–
By incorporating data and structural interdependence functions, RPN 2 explicitly models data interdependence via new component functions in its architecture, enabling it to encompass a broader range of contemporary dominant backbone models within its canonical representation.
- A compression method for object detection network for edge devices–
Through unstructured global pruning of the network and using the pruning network of the previous stage to sparsely distill the pruning network of the current stage, while maintaining the network sparsity, the problem of insufficient computing power of edge computing equipment is solved.
- Self-Adaptation Feature Attention Network with Multi-Step Fusion for Single Image Dehazing–
An efficient end-to-end self-adaptation feature attention (SAFA) network with multi-step fusion that can adaptively expand the receptive field to obtain the key structure information in space and extract more comprehensive and accurate features.
- –
- –
- –
- –
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.