Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks138 from public data
- DCSFN: Deep Cross-scale Fusion Network for Single Image Rain Removal138
This paper explores the cross-scale manner between networks and inner-scale fusion operation to solve the image rain removal task and proposes a multi-sub-networks structure, which outperforms over the state-of-the-art methods.
- GELU-Net: A Globally Encrypted, Locally Unencrypted Deep Neural Network for Privacy-Preserved Learning126
This paper proposes a novel privacy-preserved architecture where clients can collaboratively train a deep model while preserving the privacy of each client’s data by carefully partition a deep neural network to two non-colluding parties.
- Deterministic Learning Theory117
Deterministic Learning Theory for Identification, Recognition, and Control presents a unified conceptual framework for knowledge acquisition, representation, and knowledge utilization in uncertain dynamic environments and represents a new model of information processing, i.e. a model of dynamical parallel distributed processing (DPDP).
- Correlation Matching Transformation Transformers for UHD Image Restoration102
Experimental results show that the UHDformer reduces about ninety-seven percent model sizes compared with most state-of-the-art methods while significantly improving performance under different training sets on 3 UHD image restoration tasks, including low-light image enhancement, image dehazing, and image deblurring.
- Joint Self-Attention and Scale-Aggregation for Self-Calibrated Deraining Network82
An effective algorithm, called JDNet, is proposed to solve the single image deraining problem and conduct the segmentation and detection task for applications and has better deraining results both on real-world and synthetic datasets.
- 75
- Command filter based adaptive neural trajectory tracking control of an underactuated underwater vehicle in three-dimensional space75
The stability of the control scheme proposed in this paper is proved in the sense of Lyapunov stability theory after comparative simulations were carried out to illustrate the effectiveness and good robustness of the proposed tracking controller.
- 56
- 50
- 50
- 41
- 41
- 39
- 37
- S-Box Design Based on 2D Multiple Collapse Chaotic Map and Their Application in Image Encryption36
A new chaotic S-box was constructed based on 2D-MCCM, and a diffusion method was designed based on the S- box, which improved security and efficiency and a new image encryption algorithm was proposed.
- Physical Model Guided Deep Image Deraining35
A novel network based on physical model guided learning for single image deraining, which consists of three sub-nets: rain streaks network, rain-free network, and guide-learning network, which outperforms the state-of-the-art deraining methods.
- Online-Updated High-Order Collaborative Networks for Single Image Deraining32
A high-order collaborative network with multi-scale compact constraints and a bidirectional scale-content similarity mining module to exploit features from deep networks externally and internally for rain streaks removal and an online-update learning approach, which uses real-world rainy images to fine-tune the network and update the deraining results in a self-supervised manner.
- 31
- 30
- 29
- 28
- Intelligent Grading of Tobacco Leaves Using an Improved Bilinear Convolutional Neural Network27
This work developed assembly line equipment and an RGB image classification method for tobacco grading that uses a pyramid structure, attention mechanism, and angle decision loss function to improve the bilinear convolutional neural network (a fine-grained visual categorization framework) for Tobacco grading.
- 26
- 26
- Crypto primitive of MOCVD MoS2 transistors for highly secured physical unclonable functions26
The generated PUF keys exhibit good randomness and uniqueness, providing a possibility for harvesting highly secured PUF devices with two-dimensional materials.
- DreamVideo: High-Fidelity Image-to-Video Generation with Image Retention and Text Guidance25
A high-fidelity image-to-video generation method by devising a frame retention branch based on a pre-trained video diffusion model, named DreamVideo, which outperforms the state-of-the-art method and can generate high-quality videos of different actions by providing varying prompt texts.
- 25
- Removing multiple types of noise of distributed acoustic sensing seismic data using attention-guided denoising convolutional neural network24
A Denoising network named attention-guided denoising convolutional neural network (ADNet), which uniquely construct a large number of complex forward models for pure seismic data training set to enhance the network suitability and reduces computational cost and memory consumption.
- 24
- Weakly supervised single image dehazing24
This paper proposes a novel weakly supervised network based on the multi-level multi-scale block that reduces the constraint on the training data and automatically estimates the transmission map and the atmospheric light as well as the intermediate haze-free image without using any realistic Transmission map and atmosphericLight as supervision.
- 24
- 23
- 23
- 22
- Ultra-High-Definition Image Restoration: New Benchmarks and a Dual Interaction Prior-Driven Solution19
This paper constructs UHD snow and rain benchmarks, named UHD-Snow and UHD-Rain, and proposes an effective UHD image restoration solution by considering gradient and normal priors in model design, thanks to these priors’ spatial and detail contributions.
- Robust layer segmentation of esophageal OCT images based on graph search using edge-enhanced weights18
Experiments from guinea pigs demonstrate that the EEGS framework is more robust and more accurate than the current segmentation method and can be potentially useful for the early detection of esophageal diseases.
- 17
- Single image deraining via deep shared pyramid network16
An effective deep neural network is developed, where the deep models at different feature pyramid levels share the same weight parameters, and a multi-stream dilation convolution is developed to deal with complex rainy streaks.
- 15
- Developing a weakly supervised deep learning framework for breast cancer diagnosis with HR status based on mammography images15
A weakly supervised deep learning framework BSNet was proposed which detected breast cancer with HR status and benign tumors and high-dimensional mammography of breast cancer subtypes, which helped inform early management options.
- 15
- FWB-Net: Front White Balance Network for Color Shift Correction in Single Image Dehazing Via Atmospheric Light Estimation15
A new non-homogeneous atmospheric scattering model (NH-ASM) is proposed for improving image modeling of hazy images taken under complex illumination conditions and an end-to-end CNN-based color-shift-restraining dehazing network is developed, termed as FWB-Net.
- A Handwritten Chinese Character Recognition based on Convolutional Neural Network and Median Filtering15
A model based on CNN to deal with matters of handwritten Chinese character recognition is proposed, different with conventional recognition system, input images are preprocessed by median filtering to smooth and reduce noise and has a good performance at real-time test.
- Pyramid fully residual network for single image de-raining15
Comprehensive experimental results show that the PFR-Net outperforms the state-of-the-art single de-raining methods with a big margin on Rain100H, Rain100L and Rain1200 datasets.
- 14
- Single image deraining via nonlocal squeeze-and-excitation enhancing network14
A novel deep network for single image de-raining consisting of three stages, including encoder stage, Dense Non-Local Residual Block (DNLRB) stage, and decoder stage, that adopts squeeze-and-excitation enhancing on feature maps in each convolution layer for capturing spatial contextual information.
- Single image deraining via deep pyramid network with spatial contextual information aggregation14
This paper proposes a novel deep learning based deraining method that learns multi-scale information both from the perspectives of kernels and feature maps, respectively, by designing spatial contextual information aggregation module and pyramid network module.
- Learning a multi-level guided residual network for single image deraining14
Quantitative and qualitative experimental results demonstrate the superiority of the proposed end-to-end deep learning based deraining method compared with several state-of-the-art deraining methods.
- Assessment of flight fatigue using heart rate variability and machine learning approaches13
Findings indicate that a LightGBM model trained on 12 selected HRV features and one respiratory indicator can accurately categorize flight fatigue into three levels, suggesting that this approach holds potential for reducing fatigue-related flight accidents.
- 13
- 13
- 13
- Neural Discrimination-Prompted Transformers for Efficient UHD Image Restoration and Enhancement12
A super-resolution-guided reconstruction approach, which is guided by super-resolving low-resolution features to facilitate final UHD image restoration and enhancement and enhances the quality of restored images.
- 12
- 12
- From Coarse to Fine: A Stage-Wise Deraining Net12
The proposed method is motivated by the idea that an effective deraining algorithm should have the ability to remove various remaining rain streaks in a repeated way, and design the deraining network in a coarse-to-fine manner that is multi-stage processing procedure and the parameters are shared in each stage.
- 12
- 11
- Phrase Grounding-Based Style Transfer for Single-Domain Generalized Object Detection11
This paper proposes a novel phrase-grounding-based style transfer (PGST) approach for single-domain generalized object detection that significantly improves mean average precision (mAP) and outperforms or matches the performance of domain-adaptive object detection methods, which require target domain data for training, in several challenging scenarios.
- 11
- 11
- Close the Gap between Deep Learning and Mobile Intelligence by Incorporating Training in the Loop11
A new framework is designed and implemented to enlarge sample space via data paring and learn a deep metric under the privacy, memory and computational constraints and indicates that training consumes lower energy than watching videos so can be scheduled intermittently on mobile devices.
- A new filter collaborative state transition algorithm for two-objective dynamic reactive power optimization11
A Filter Collaborative State Transition Algorithm (FCSTA) is presented for solving DRPO problems and the effectiveness of the proposed algorithms are compared with different approaches in the literature.
- 10
- 10
- 9
- 9
- Real-world image denoising via efficient diffusion model with controllable noise generation9
This work presents a general denoising method inspired by diffusion models, utilizing simple convolutional neural networks such as UNet, and achieves denoising performance comparable to that of the transformer architecture.
- 9
- 9
- Application of SOM Neural Network in Lithology Recognition9
A identification model was established to recognize lithology in northern Qaidam basin according to the advantage of the recognition of linear distribution with SOM network, and it apparently displayed more efficient than other traditional method.
- Iterative Optimal Attention and Local Model for Single Image Rain Streak Removal8
Extensive experiments validate that the proposed EMResformer surpasses current state-of-the-art single image rain streak removal methods on both synthetic and real-world datasets, achieving an improved balance between model complexity and single image deraining performance.
- Non-Contact Blood Pressure Monitoring Using Radar Signals: A Dual-Stage Deep Learning Network8
A hierarchical neural framework that synergizes spatial and temporal feature learning for radar-driven, contactless BP monitoring, and closely approximates real-time, beat-to-beat monitoring, representing a pivotal breakthrough in non-contact BP monitoring.
- Progressive Local and Non-Local Interactive Networks with Deeply Discriminative Training for Image Deraining8
A progressive local and non-local interactive network with multi-scale cross-content deeply discriminative learning to solve image deraining and outperforms state-of-the-art methods on five public synthetic datasets and one real-world data.
- 8
- Semantic-Guided Global–Local Collaborative Networks for Lightweight Image Super-Resolution7
This article proposes a semantic-guided global-local collaborative network (SGGLC-Net) for lightweight SISR that seamlessly integrates the semantic priors into the super-resolution network, enabling the network to more adeptly capture and utilize semantic priors, thereby enhancing image details.
- 7
- 7
- Recent Advances in Integrating OWL and Rules (Technical Communication)7
This paper takes OWL 2 DL as a starting point and pursue the question of how features of rule-based formalisms can be added without jeopardizing decidability, and reports on incorporating the closed world assumption and on reasoning algorithms.
- GTIGNet: Global Topology Interaction Graphormer Network for 3D hand pose estimation6
The Global Topology Interaction Graphormer Network (GTIGNet) is introduced, a novel deep learning architecture designed to improve 3D hand pose estimation that incorporates a Context-Aware Attention Block within the 2D pose estimator to enhance the extraction of multi-scale features, yielding more accurate 2D joint heatmaps to support the task that followed.
- PercepLIE: A New Path to Perceptual Low-Light Image Enhancement6
By combining the VQGAN with MDC and GCA within a stage-wise training mechanism, the PercepLIE method generates images with finer details and natural colors and achieves favorable performance on both synthetic and real-world datasets in terms of perceptual quality metrics.
- 6
- 6
- 6
- 6
- Structural Prior Guided Generative Adversarial Transformers for Low-Light Image Enhancement5
A new structural prior guided adversarial learning method is developed by building the skip connections between the generator and discriminators so that the discriminators can better discriminate between real and fake features.
- 5
- Single image rain streak removal via layer similarity prior5
This is the first paper to investigate the effectiveness of exploiting the layer similarity prior and the multi-dense-short-connection and Quantitative and qualitative experimental results demonstrate that the proposed method outperforms other state-of-the-art methods with the least parameters.
- Single image rain removal via densely connected contextual and semantic correlation net5
This work designs a deraining unit that employs dilation convolution and squeeze-and-excitation operations, respectively, to obtain more spatial contextual information and semantic correlation and interconnect the deraining units by dense connections that can maximize the information flow along features from different levels and make them be associated.
- 5
- 5
- 5
- A Single Image Enhancement Technique Using Dark Channel Prior4
Extensive experiments show that the improved algorithm can not only reduce the time complexity, but also effectively improve the edge effect, and can also solve the problem of failure of dark channel prior.
- Densely connected multi-scale de-raining net4
Quantitative and qualitative experimental results demonstrate the superiority of proposed method compared with several state-of-the-art de-raining methods, while the parameters of the proposed method are greatly reduced that benefits from the proposed shared parameters strategy at different scales.
- 4
- Kernel Semi-supervised Extreme Learning Machine Applied in Urban Traffic Congestion Evaluation4
Kernel-SSELM model is used to train the traffic congestion evaluation framework, with both small-scale labeled data and large-scale unlabeled data, and shows the evaluation system can precisely reflect the traffic condition.
- 3
- 3
- 2
- Multi-Attention Based Visual-Semantic Interaction for Few-Shot Learning2
A novel Multi-Attention based Visual-Semantic Interaction (MAVSI) approach for FSL that utilizes spatial and channel attention mechanisms to effectively select discriminative visual features for the support set based on its ground-truth semantics while using all the support set semantics for each query set sample.
- 2
- Mechanism-Parameters Optimization of a Reconfigurable Tracked Mobile Modular Deep-Sea Rover ROV2
A method to obtain optimized mechanism-parameters for the R-ROV considering the underwater environment, which has been calculated through simulation is proposed and the results are reported in detail.
- A Multi-Objective Optimization of the Reactor Power Plant2
The study shows that the component mathematical models are reliable for the optimization process, the distribution of the non-dominated solution is decided by the steam generator secondary pressure.
- 1
- Solving non-cyclic dynamic multi-objective optimization problems via GRU prediction and multi-information hybrid exploration1
The proposed algorithm GPMHE has shown robust adaptation to dynamic environmental changes and compared with five advanced algorithms on the non-cyclic dynamic benchmark test set (NCD) has shown robust adaptation to dynamic environmental changes.
- Discrete state transfer algorithm based on deep reinforcement learning for solving flexible job shop scheduling problem1
The comparison results show that the proposed DSTA-PPO algorithm can solve the examples in most examples with a shorter completion time, and effectively solve the flexible job shop scheduling problem.
- 1
- Answer Path at NTCIR-7 CCLQA Track1
This is the first time that this group participate NTCIR and Answer Path is a brand new system, and the main components are Question Analyzer, Passage Retrieval and Answer Extractor, and Sentence Resemble Model and Answer Filtering.
- 1
- MambaLIE: Scene Light Intensity-Boosted Low-Light Image Enhancement with State Space Model–
This work proposes MambaLIE, a Scene Light Intensity-Boosted Low-Light Image Enhancement method based on a State Space Model (SSM), which outperforms state-of-the-art CNN-based and Transformer-based LIE methods on four widely used synthetic benchmarks and five publicly available real-world benchmarks in terms of accuracy, speed, and model size.
- Multi-Modal Object Re-Identification With Prompt-S6 and Semantic-Aware Knowledge Guidance–
A novel multi-modal ReID framework built upon Prompt-S6 (PS6) and semantic-aware knowledge guidance, which generates more robust multi-modal representations under complex scenarios and leverages semantic priors from the segmentation foundation models.
- Multi-Modal Object Re-Identification with Dual Semantic Guidance and Global-Local Mutual Modulation–
A robust multi-modal ReID framework with dual semantic guidance and global-local mutual modulation, which mainly consists of three key components, namely the Text-Semantic Injector (TSI), the Masked Global-Local Modulator (MGLM), and the Hierarchical MoE Fusion (HMF).
- Degradation-Aware Prompt Learning With Cross-Modal Compensation for Adverse Weather Removal–
A Degradation-Aware Cross-Modal Prompt Compensation Network that leverages cross-modal degradation cues from a pre-trained vision-language model to condition restoration features within a unified backbone, achieving robust and perceptually consistent restoration across diverse weather conditions.
- –
- –
- Perspective on Artificial Intelligence for Security–
This special issue on AI Security in the Transactions on Artificial Intelligence brings together contemporary research on the threats, defenses, and emerging paradigms associated with secure AI deployment.
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- A Discrete Biogeography-Based Optimization for Solving Tomato Planting Planning–
Results showed that tomato planting planning scheme calculated by the proposed algorithm can realize the balance supplement of tomato materials effectively.
- –
- –
- –
- –
- –
- Reasoning Approaches for Nominal Schemas–
This paper summarizes several reasoning approaches recently devised to reason over nominal schemas, which allows DL languages to express arbitrary DL-safe rules in their native syntax.
- Skin detection based on maximal class separability in new color space–
The experimental results indicate that the new color spaces based on maximal class separability performs better than traditional color spaces and the illustration of skin segmentation in color images also verifies the effectivenesses of both new color space model and assembled model.
- –
- –
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 2026-10-15. 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.