Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks211 from public data
- Frequency-Aware Feature Fusion for Dense Image Prediction325
Comprehensive visualization and quantitative analysis demonstrate that FreqFusion effectively improves feature consistency and sharpens object boundaries, and extensive experiments across various dense prediction tasks confirm its effectiveness.
- Partial FC: Training 10 Million Identities on a Single Machine300
This work theoretically analyze the upper limit of model parallelism in face recognition and proposes a load-balanced sparse distributed classification training method, Partial FC, which is capable of using a machine with only 8 Nvidia Tesla V100 GPUs to implement training on a face recognition data set with up to 29 million IDs.
- Infrared Small Target Detection with Scale and Location Sensitivity292
A novel Scale and Location Sensitive (SLS) loss is proposed to handle the limitations of existing losses and a simple Multi-Scale Head to the plain U-Net (MSHNet) is designed, which outperforms existing state-of-the-art methods by a large margin.
- A Physics-Based Noise Formation Model for Extreme Low-Light Raw Denoising280
A highly accurate noise formation model based on the characteristics of CMOS photosensors is presented, thereby enabling us to synthesize realistic samples that better match the physics of image formation process.
- 3-D Quasi-Recurrent Neural Network for Hyperspectral Image Denoising247
An alternating directional 3-D quasi-recurrent neural network for hyperspectral image (HSI) denoising, which can effectively embed the domain knowledge—structural spatiospectral correlation and global correlation along spectrum (GCS).
- Single Image Reflection Removal Exploiting Misaligned Training Data and Network Enhancements201
Experimental results collectively show that the method outperforms the state-of-the-art with aligned data, and that significant improvements are possible when using additional misaligned data.
- 201
- 192
- Physics-based Noise Modeling for Extreme Low-light Photography188
Extensive experiments on multiple low-light denoising datasets show that a deep neural network trained with the proposed noise formation model can reach surprisingly-high accuracy, opening a new door to real-world extreme low- light photography.
- Deep-Learning-Based Semantic Segmentation of Remote Sensing Images: A Survey173
A comprehensive review of semantic segmentation models based on deep learning in SSRSI, including semisupervised and weakly-supervised SSRSI, unsupervised domain adaption in SSRSI, multimodal data-fusion-based SSRSI, and pretrained models for SSRSI are reviewed.
- 173
- 141
- Spectral Enhanced Rectangle Transformer for Hyperspectral Image Denoising134
This paper proposes a spectral enhanced rectangle Trans-former, driving it to explore the nonlocal spatial similarity and global spectral low-rank property of HSIs, and designs a spectral enhancement module that is capable of extracting global underlying low- rank property of spatial-spectral cubes to suppress noise.
- Instance Segmentation in the Dark128
This work takes a deep look at instance segmentation in the dark and introduces several techniques that substantially boost the low-light inference accuracy and discovers that high-bit-depth RAW images can better preserve richer scene information in low-light conditions compared to typical camera sRGB outputs, thus supporting the use of RAW-input algorithms.
- 126
- 118
- Frequency Dynamic Convolution for Dense Image Prediction114
Frequency Dynamic Convolution is introduced, a novel approach that mitigates limitations by learning a fixed parameter budget in the Fourier domain and seamlessly integrates into a variety of architectures, including ConvNeXt, Swin-Transformer, offering a flexible and efficient solution for modern vision tasks.
- 107
- 105
- Bidirectional 3D Quasi-Recurrent Neural Network for Hyperspectral Image Super-Resolution103
A bidirectional 3D quasi-recurrent neural network is designed that can be well generalized to remote sensing HSI datasets with limited number of training data and demonstrates improvements over state-of-the-art methods, in terms of both restoration accuracy and visual quality.
- 101
- LG-BPN: Local and Global Blind-Patch Network for Self-Supervised Real-World Denoising99
A novel method called LG-BPN is presented, which takes the spatial correlation statistic into the authors' network design for local detail restoration, and also brings the long-range dependencies modeling ability to previously CNN-based BSN methods.
- 98
- 90
- 90
- Pixel Adaptive Deep Unfolding Transformer for Hyperspectral Image Reconstruction86
A Pixel Adaptive Deep Unfolding Transformer (PADUT) for HSI reconstruction is proposed, and the Non-local Spectral Transformer (NST) is introduced to emphasize the 3D characteristics of HSI for recovering.
- 86
- 83
- Deep plug-and-play prior for hyperspectral image restoration81
This paper develops a new deep HSI denoiser leveraging gated recurrent convolution units, short- and long-term skip connections, and an augmented noise level map to better exploit the abundant spatio-spectral information within HSIs and demonstrates that this approach often achieves superior performance, which is competitive with or even better than the state-of-the-art on each task.
- Highly specific noninvasive photoacoustic and positron emission tomography of brain plaque with functionalized croconium dye labeled by a radiotracer77
An NIR ultrahigh absorbing croconium dye for amyloid (CDA) was designed and synthesized and described in detail in a report published in 2016.
- 76
- Generative Adversarial Network–based Noncontrast CT Angiography for Aorta and Carotid Arteries71
A GAN-based model that synthesizes neck and abdominal CTA-like images without the use of ICAs shows promise in vascular diagnosis compared with real CTA images.
- Automated classification of software change messages by semi-supervised Latent Dirichlet Allocation71
This work presents a novel automatic change message classification method characterized by semi-supervised topic semantic analysis that automatically classifies most of the change messages which record the cause of the software change and is applicable to cross-project analysis of software change messages.
- Intensity and Stationarity Analysis of Land Use Change Based on CART Algorithm69
The change of land use during 1995–2015 indicated that Tianjin experienced rapid urban development with the area of urban land increased by about 7.5%.
- 67
- LocalTrans: A Multiscale Local Transformer Network for Cross-Resolution Homography Estimation65
Experiments show that the proposed network outperforms existing state-of-the-art feature-based and deep-learning-based homography estimation methods, and is able to accurately align images under 10× resolution gap.
- Adaptive Spatial-Spectral Dictionary Learning for Hyperspectral Image Restoration65
An effective model for hyperspectral image (HSI) restoration, specifically image denoising and super-resolution is proposed, which can be effectively solved by an iterative numerical algorithm with parameters that are adaptively adjusted for different clusters and different noise levels.
- Joint Camera Spectral Sensitivity Selection and Hyperspectral Image Recovery63
An efficient convolutional neural network (CNN) based method, which can jointly select the optimal CSS from a candidate dataset and learn a mapping to recover HSI from a single RGB image captured with this algorithmically selected camera.
- 59
- GPS-Net: Graph-based Photometric Stereo Network59
A Graph-based Photometric Stereo Network, which proposes the Unstructured Feature Extraction Layer to connect an arbitrary number of input image-light pairs into graph structures, and introduces Structure-aware Graph Convolution to balance the input data by appropriately weighting shadows and specular highlights.
- 56
- 51
- 50
- 50
- 3D magnetic resonance image denoising using low-rank tensor approximation48
Experimental results on the proposed noise reduction method show that the proposed method can provide substantial improvements over the current state-of-the-art image denoising methods in terms of both objective metric and subjective visual quality.
- Dynamics and adaptive synchronization of the energy resource system48
Based on Lyapunov stability theory, an adaptive control law is derived to make the states of two energy resource systems with different unknown system parameters asymptotically synchronized.
- RawHDR: High Dynamic Range Image Reconstruction from a Single Raw Image46
The core idea of this work is to incorporate more informative Raw sensor data to generate HDR images, aiming to recover scene information in hard regions (the darkest and brightest areas of an HDR scene), and proposes a model tailor-made for Raw images, harnessing the unique features of Raw data to facilitate the Raw-to-HDR mapping.
- 43
- Hybrid Spectral Denoising Transformer with Guided Attention42
A hybrid approach that combines the advantages of both models with a Spatial-Spectral Separable Convolution (S3Conv), Guided Spectral Self-Attention (GSSA), and Self-Modulated Feed-Forward Network (SM-FFN) for hyperspectral image denoising.
- 42
- 41
- End-to-End Video Text Spotting with Transformer39
This paper proposes a simple, yet effective end-to-end trainable video text DEtection, Tracking, and Recognition framework (TransDeTR), which views the VTS task as a direct long-range temporal modeling problem.
- Guided Hyperspectral Image Denoising with Realistic Data39
The extensive experimental results show that a network learned with only synthetic data generated by the noise model performs as well as it is learned with paired real data, and the guided HSI denoising network outperforms state-of-the-art methods under both quantitative metrics and visual quality.
- Investigating and improving log parsing in practice39
An improved log parser named Drain+ is proposed based on a state-of-the-art log parser Drain, which includes two innovative components to address two challenges for log parsing in practice: a statistical-based separators generation component, which generates separators automatically for log message splitting, and a candidate event template merging component which merges the candidate event templates by a template similarity method.
- 39
- Hyperspectral Image Super Resolution With Real Unaligned RGB Guidance38
This article proposes an HSI fusion network (HSIFN) with heterogeneous feature extractions, multistage feature alignments, and attentive feature fusion, and shows that this method obtains a clear improvement over existing single-image and fusion-based super-resolution methods on quantitative assessment as well as visual comparison.
- Suppress and Rebalance: Towards Generalized Multi-Modal Face Anti-Spoofing38
The Uncertainty-Guided Cross-Adapter (U-Adapter) is proposed to recognize unreliably detected regions within each modality and suppress the impact of unreliable regions on other modal-ities, and a Rebalanced Modality Gradient Modulation (ReGrad) strategy to rebal-ance the convergence speed of all modalities by adaptively adjusting their gradients is proposed.
- When Semantic Segmentation Meets Frequency Aliasing38
A comprehensive analysis of hard pixel errors, categorizing them into three types: false responses, merging mistakes, and displacements is conducted, revealing a quantitative association between hard pixels and aliasing, which is distortion caused by the overlapping of frequency components in the Fourier domain during downsampling.
- An empirical study of the impact of log parsers on the performance of log-based anomaly detection36
A comprehensively empirical study to investigate the impact of six state-of-the-art log parsers belonging to four categories (including heuristic-based, frequency- based, clustering-based; and subsequence-based) on six state of theart log-based anomaly detection methods (including machine-learning-based and deep- learning-based methods).
- Transformer Based Pluralistic Image Completion With Reduced Information Loss35
A patch-based auto-encoder that directly takes features from the P-VQVAE encoder as input without any quantization and only regards the quantized tokens as prediction targets is designed, which greatly outperforms existing transformer based methods on image fidelity.
- 35
- 34
- 33
- 33
- Distilling Monocular Foundation Model for Fine-grained Depth Completion32
A two-stage knowledge distillation framework that leverages powerful monocular foundation models to provide dense supervision for depth completion and enables depth completion models to harness the strengths of monocular foundation models to harness the strengths of monocular foundation models.
- EventHDR: From Event to High-Speed HDR Videos and Beyond32
This paper presents a recurrent convolutional neural network that reconstruct high-speed HDR videos from event sequences, with a key frame guidance to prevent potential error accumulation caused by the sparse event data.
- Low-rank Bayesian tensor factorization for hyperspectral image denoising31
This paper designs a hierarchical probabilistic model based on Bayesian tensor factorization to capture the inherent spatial-spectral correlation of HSI, which can be effectively solved under the variational Bayesian framework.
- Deep Spatial Adaptive Network for Real Image Demosaicing29
A deep spatial adaptive network (SANet) for real image demosaicing, which can adaptively learn the nonlinear mapping function for different locations and the weights of spatial adaptive convolution layer are generated by the pattern information in the receptive filed.
- Estimating Fine-Grained Noise Model via Contrastive Learning29
This work combines both noise modeling and estimation, and proposes an innovative noise model estimation and noise synthesis pipeline for realistic noisy image generation and endows deep denoisers with competitive performances with state-of-the-art real noise modeling methods.
- Disentangled Face Attribute Editing via Instance-Aware Latent Space Search29
A novel framework (IALS) that performs Instance-Aware Latent-Space Search to find semantic directions for disentangled attribute editing and outperforms state-of-the-art methods proposed recently by a wide margin.
- Denoising Diffusion Semantic Segmentation with Mask Prior Modeling28
This paper proposes to ameliorate the semantic segmentation quality of existing discriminative approaches with a mask prior modeled by a recently-developed denoising diffusion generative model, and develops a unified architecture that adapts diffusion models for mask prior modeling.
- 28
- 27
- 26
- 24
- Multi-stimuli-responsive cyanostilbene derivatives: Their fluorescent and mechanochromic properties, and potential application in water sensing and anti-counterfeiting24
The potential in the fields of water sensing and anti-counterfeiting for these two multi-stimuli-responsive compounds TPE-CNTPA and PH-CNTP have enormous potential for application in the field of encryption anti-counterfeiting.
- Simultaneous hyperspectral image super-resolution and geometric alignment with a hybrid camera system24
Experimental results show that the proposed method outperforms the current state-of-the-art HSI super-resolution methods with a misaligned hybrid camera system in terms of both objective metric and subjective visual quality.
- PIRM2018 Challenge on Spectral Image Super-Resolution: Methods and Results24
The Perceptual Image Restoration and Manipulation (PIRM) workshop challenge on spectral image super-resolution is described, the methods used by the participants, summarise the results and discuss their rankings.
- Siamese-DETR for Generic Multi-Object Tracking23
This paper focuses on TIMOT and proposes a simple but effective method, Siamese-DETR, which surpasses existing MOT methods on GMOT-40 dataset by a large margin and leverages the inherent object queries in DETR variants.
- 23
- 23
- 23
- Cross-modal dynamic convolution for multi-modal emotion recognition23
This work presents cross-modal dynamic convolution, a method that can achieve competitive performance compared to previous works while being more efficient, and is easy to stack, enabling it to model long-range cross- modal temporal interactions.
- 22
- 22
- 22
- 22
- 22
- Safeguarding Medical Image Segmentation Datasets against Unauthorized Training via Contour- and Texture-Aware Perturbations21
An Unlearnable Medical image generation method, termed UMed, which integrates the prior knowledge of MIS by injecting contour- and texture-aware perturbations to protect images.
- 21
- Mixed Attention Network for Hyperspectral Image Denoising20
An attentive skip-connection that adaptively controls the propor-tion of the low- and high-level spatial-spectral features from the encoder and decoder to better enhance the aggregated features is proposed.
- 20
- Efficient Hybrid Supervision for Instance Segmentation in Aerial Images20
This work designs an ancillary segmentation model that can locate object pixel-wisely instead of relying on horizontal bounding box prediction, which has better adaptability to arbitrary oriented objects.
- Community Detection Based on Regularized Semi-Nonnegative Matrix Tri-Factorization in Signed Networks20
A framework based on regularized semi-nonnegative matrix tri-factorization is proposed which maps the signed network from high-dimensional space to low- dimensional space, such that the communities of thesigned network can be derived.
- Interreflection Removal Using Fluorescence20
A method for removing interreflections from a single image using fluorescence is proposed from a bispectral observation of reflective and fluorescent components recorded in distinct color channels and shows how it improves an existing photometric stereo method in shape recovery.
- 19
- Recurrent Self-Supervised Video Denoising with Denser Receptive Field19
RDRF is proposed for selfsupervised video denoising, which not only fully exploits both the reference and neighbor frames with a denser receptive field, but also better leverages the temporal information from both local and distant neighbor features.
- Local-enhanced transformer for single-pixel imaging19
This Letter proposes a high-quality under-sampled SPI method based on a novel, to the best of the authors' knowledge, local-enhanced transformer that is not only good at capturing global dependencies of SPI measurements, but also has the capability to model local dependencies.
- 18
- Translation of Aerial Image Into Digital Map via Discriminative Segmentation and Creative Generation18
- 17
- 17
- 17
- Cross-modal context-gated convolution for multi-modal sentiment analysis17
This work proposes cross-modal context-gated convolution for unaligned sequences, which captures the local cross- modal interactions, dealing with the misalignment while reducing the effect of unrelated information.
- Flow-Anything: Learning Real-World Optical Flow Estimation From Large-Scale Single-View Images16
The benefits of generating optical flow training data from large-scale real-world images, outperforming the most advanced unsupervised methods and supervised methods on synthetic datasets are demonstrated.
- 16
- MRI Denoising Using Low Rank Prior and Sparse Gradient Prior16
The experimental results show that the proposed algorithm outperforms many classical MRI denoising methods, such as unbiased nonlocal means (UNLM) filtering and block-matching and 3D (BM3D) filtering in both visual and numerical results.
- Spectral reflectance recovery using optimal illuminations15
This paper designs illumination optimization layer to optimally multiplex illumination spectra in a given dataset or to design the optimal one under physical restrictions and develops the nonlinear representation for spectral reflectance in a data-driven way and jointly optimize illuminations under this representation in a CNN-based end-to-end architecture.
- MPI-Flow: Learning Realistic Optical Flow with Multiplane Images14
This work constructs a layered depth representation, known as multiplane images (MPI), from single-view images, and presents an independent object motion module that can separate the camera and dynamic object motion in MPI, to ensure the realism of motion.
- 14
- Weakly-supervised Semantic Segmentation in Cityscape via Hyperspectral Image14
A weakly-supervised HSI semantic segmentation framework is introduced, which utilizes spectral information to improve the coarse labels to a finer degree and can obtain highly competitive labels and even have higher edge fineness than artificial fine labels in some classes.
- 14
- Image restoration from patch-based compressed sensing measurement14
This paper presents a non-iterative image reconstruction method from patch-based compressively sensed random measurement that outperforms state-of-the-art compressive sensing (CS) reconstruction methods with patch- based CS measurement and applies the de-block process in this method to JPEG compression artifacts removal.
- 13
- 13
- Dynamic proximal unrolling network for compressive imaging13
Experimental results demonstrate that the proposed DPUNet can effectively handle multiple compressive imaging modalities under varying sampling ratios and noise levels via only one trained model, and outperform the state-of-the-art approaches.
- Polydimethylsiloxane (PDMS)-Based Flexible Optical Electrodes with Conductive Composite Hydrogels Integrated Probe for Optogenetics13
The flexible penetrating optical electrode developed may be used for long-term in vivo optogenetics studies and showed a lower impedance and higher charge capacity after PEDOT/PSS modification.
- 13
- Real Noise Decoupling for Hyperspectral Image Denoising12
A multi-stage noise-decoupling framework that decomposes complex noise into explicitly modeled and implicitly modeled components, which reduces the complexity of noise and enhances the learnability of HSI denoising methods when applied to real paired data is proposed.
- Consistency-Aware Map Generation at Multiple Zoom Levels Using Aerial Image11
A novel multilevel map (MLM) generation framework that can automatically generate accurate and consistent maps with multiple zoom levels from aerial images is proposed, which outperforms the previous state-of-the-art map generation methods on both comprehensive quantitative metrics and perceptual quality.
- 11
- 10
- An Effective Network with ConvLSTM for Low-Light Image Enhancement10
A convolutional Long Short-Term Memory (ConvLSTM) network based method to directly restore a normal image from a low-light image, which can be learned in an end-to-end way and outperforms the state-of-the-art traditional and deep learning based methods.
- Hyperspectral image super‐resolution under misaligned hybrid camera system10
Experimental results show that the proposed method outperforms the state-of-the-art hyperspectral image super-resolution methods under a misaligned hybrid camera system in terms of both objective metric and subjective visual quality.
- 9
- Relation-Guided Adversarial Learning for Data-Free Knowledge Transfer9
A novel Relation-Guided Adversarial Learning method with triplet losses, which solves the homogeneity problem from two aspects and shows significant improvement over previous state-of-the-art methods in accuracy and data efficiency.
- 9
- 9
- 9
- 8
- 8
- 8
- CSR-Net: Camera Spectral Response Network for Dimensionality Reduction and Classification in Hyperspectral Imagery8
The experimental results of four popular public hyperspectral datasets show that the proposed CSR-Net method, with only a few image bands, outperforms state-of-the-art HSI classification methods which utilize the full spectral bands of images.
- 7
- 7
- 7
- RobuSTereo: Robust Zero-Shot Stereo Matching under Adverse Weather6
RobuSTereo is proposed, a novel framework that enhances the zeroshot generalization of stereo matching models under adverse weather by addressing both data scarcity and feature extraction challenges and significantly improves the robustness and generalization of stereo matching models across diverse adverse weather scenarios.
- 6
- Mono2Stereo: Monocular Knowledge Transfer for Enhanced Stereo Matching6
This work proposes leveraging monocular knowledge transfer to enhance stereo matching, namely Mono2Stereo, with a two-stage training process, comprising synthetic data pre-training and real-world data fine-tuning.
- Learning Dense Feature Matching via Lifting Single 2D Image to 3D Space5
A novel two-stage framework that lifts 2D images to 3D space, named as Lift to Match (L2M), taking full advantage of large-scale and diverse single-view images and achieving generalization across diverse domains is proposed.
- 5
- 5
- 5
- Global relative position space based pooling for fine-grained vehicle recognition5
A global relative position space based pooling (GRPSP) method to replace the absolute position information withglobal relative position information, by which the discriminative features can be well aligned and thereby the intra-class variation is reduced effectively.
- Direct and Global Component Separation from a Single Image Using Basis Representation5
This work presents a novel approach for separating direct and global components of a scene in full spatial resolution from a single captured image, which employs linear basis representation to approximate direct andglobal components.
- Deep‐Learning Aided Atomic‐Scale Observation of Anisotropic Melting of the Charge Density Wave in TaS 24
AI-augmented scanning transmission electron microscopy combined with differential phase contrast (DPC) imaging is employed to resolve the anisotropic melting dynamics of the nearly commensurate CDW (NC-CDW) phase in 1T-TaS2 and establish a methodology for atomic-scale manipulation of CDW states.
- 4
- 4
- Event-guided Video Clip Generation from Blurry Images4
An end-to-end convolutional neural network based method under the local and global constraints of events to restore clear, sharp intensity frames through collaborative learning from a blurry image and its associated event streams is presented.
- 4
- 4
- GPU Assisted Towards Real-Time Reconstruction for Dual-Camera Compressive Hyperspectral Imaging4
A GPU assisted towards real-time reconstruction framework for DCCHI is proposed, leveraging the fast convergence rate of the alternative direction multiplier method, and a reformative reconstruction algorithm which can achieve a fast converge rate.
- 4
- 4
- 3
- 3
- Semantic-Attention Enhanced DSC-Transformer for Lymph Node Ultrasound Classification and Remote Diagnostics3
A novel Semantic-Attention Enhanced Dynamic Swin Convolutional Block Attention Module(CBAM) Transformer (DSC-Transformer) for lymph node ultrasound image classification, representing a significant advancement in AI-driven medical image analysis with broad implications for telehealth deployment.
- 3
- 3
- 3
- 3
- Oriented Object Detection in Remote Sensing Using an Enhanced Feature Pyramid Network3
An oriented bounding box optical remote sensing image object detection method based on an enhanced feature pyramid network, utilizing deformable convolutions and feature fusion modules to enhance the feature information of rotating targets and improve the expressive capacity of features at all levels is proposed.
- Complex Network Dynamics of Multiscroll Chaotic Attractors and Their Output-Feedback Pinning Synchronization3
A new output-feedback pinning controller is designed based on the unknown-input distributed observer to estimate the synchronization error, which has the advantage of saving the communication cost as the transmission of the observer information is not needed.
- 3
- 3
- Make U-Net Greater: An Easy-to-Embed Approach to Improve Segmentation Performance Using Hypergraph3
A plug-and-play module aggregating both local and global information (aka LGIA module) to capture the high-order relationship between nodes that are far apart is proposed which outperformed several baseline models.
- 3
- 3
- 3
- StructDiff: Structure-aware Diffusion Model for 3D Fine-grained Medical Image Synthesis2
The Fine-grained Cardiac image Synthesis framework is proposed, established on 3D template conditional diffusion model, and a Confidence-aware Adaptive Learning (CAL) strategy is proposed to facilitate the pre-training of downstream segmentation tasks.
- 2
- DSA: Deformable Segmentation Attention for Multi-Scale Fisheye Image Segmentation2
The deformable segmentation attention (DSA) module is designed, a generalpurpose architecture based on a deformable attention mechanism and a spatial pyramid architecture to improve the perception of fisheye images by traditional networks without increasing the amount of excessive computation.
- Snapshot Multiplexed Imaging Based on Compressive Sensing2
A snapshot multiplexed imaging system for wide FoV imaging that model the demultiplexing as a compressive sensing (CS) reconstruction problem and introduces two methods, one based on Total Variation (TV) constraint and the other based on sparsity constraint, to reconstruct the scene.
- 2
- [Development of early auditory and speech perception skills within one year after cochlear implantion in prelingual deaf children].2
Within one year after cochlear implantation, children develop early auditory and speech perception capabilities with time and Chinese versions of the IT-MAIS/MAIS, MESP and MPSI are useful instruments to document early auditoryand speech perception skills in children after CI implantation.
- Cryo–electron microscopy structures of human cone visual pigments1
Cryo-electron microscopy structures of the active cone opsin structures explain how amino acid substitutions fine-tune spectral sensitivity and help clarify the molecular basis of color vision deficiencies and key differences in rod versus cone activation.
- 1
- 1
- MARPO: A Reflective Policy Optimization for Multi Agent Reinforcement Learning1
MARPO consists of a reflection mechanism that leverages subsequent trajectories to enhance sample efficiency, and an asymmetric clipping mechanism that is derived from the KL divergence and dynamically adjusts the clipping range to improve training stability.
- 1
- 1
- 1
- Illumination modulation for reflective and fluorescent separation1
An optical system using off-the-shelf devices to generate high frequency illumination, which is desirable in reflective-fluorescent separation tasks, and employs the total variation regularization scheme to account for spectral-spatial correlation, which makes the method robust to noise.
- 1
- Fast HSI super resolution using linear regression1
The authors model the hyperspectral super resolution as a linear regression problem according to the fact that the imaging process is a linear transform and the inverse of this transform can be approximately estimated, as the spectra of a typical scene lie in a very low-dimensional space.
- –
- –
- –
- RDANet: Relative Degradation Aware Network for Infrared Small Target Detection–
A Relative Degradation Aware Network (RDANet) for infrared small target detection is proposed, which achieves the best performance on most evaluation metrics, while scale- and background-stratified evaluations indicate more stable behavior across target sizes and scene complexity.
- –
- MAFE R-CNN: Selecting More Samples to Learn Category-aware Features for Small Object Detection–
This paper identifies that this difficulty stems from the detectors' inability to effectively learn discriminative features for objects of small size, compounded by the complexity of selecting high-quality small object samples during training, which motivates the proposal of the Multi-Clue Assignment and Feature Enhancement R-CNN.
- –
- –
- –
- –
- –
- –
- –
- –
- –
- ProbNVS: Fast Novel View Synthesis with Learned Probability-Guided Sampling–
This work proposes to build a novel view synthesis framework based on learned MVS priors that enables general, fast and photo-realistic view synthesis simultaneously and achieves 15 to 40 times faster rendering compared to state-of-the-art baselines, with strong generalization capacity and comparable high-quality Novel view synthesis performance.
- –
- Improved Quasi-Recurrent Neural Network for Hyperspectral Image Denoising–
This paper introduces an adaptive fusion module to replace its vanilla additive skip connection to better fuse the features of the encoder and decoder and identifies several important techniques to further enhance the performance, which includes removing batch normalization, use of extra frequency loss, and learning rate warm-up.
- –
- –
- –
- –
- –
- –
- Network grouping and upgrading for communication exchange engineering–
The upgrading of communication network brought about radical changes in the mode of communication, made the comunication management highly efficient and very smart, and greatly improved the quality of communication.
- –
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 2026-10-14. 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.