Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileAcademic lineage
Possible advisorsa guess from early papers, not confirmed
- Maojun ZhangSuggested from co-authorship
Is this you? Claim this profile to confirm or dismiss it.
Works62 from public data
- Revisiting Dilated Convolution: A Simple Approach for Weakly- and Semi-Supervised Semantic Segmentation627
It is found that varying dilation rates can effectively enlarge the receptive fields of convolutional kernels and more importantly transfer the surrounding discriminative information to non-discriminative object regions, promoting the emergence of these regions in the object localization maps.
- 397
- Accelerating thrombolysis using a precision and clot-penetrating drug delivery strategy by nanoparticle-shelled microbubbles194
A porous and magnetic microbubble platform is developed to formulate tPA and this strategy not only improves the therapeutic efficacy but also accelerates the lytic rate, enabling it to be promising in time-critical thrombolytic therapy.
- 152
- Video Scene Parsing with Predictive Feature Learning117
It is experimentally proved that the learned predictive features in the model are able to significantly enhance the video parsing performance by combining with the standard image parsing network.
- 74
- 62
- A Class Imbalance Loss for Imbalanced Object Recognition50
A method for generating an imbalanced few-shot learning dataset and the effectiveness of an algorithm for model-agnostic metalearning for rapid adaptation with CI loss are presented.
- 42
- NTIRE 2022 Challenge on Stereo Image Super-Resolution: Methods and Results35
The 1st NTIRE challenge on stereo image super-resolution (restoration of rich details in a pair of low-resolution stereo images) with a focus on new solutions and results establishes a new benchmark for stereo image SR.
- Efficient Long-Short Temporal Attention network for unsupervised Video Object Segmentation28
An efficient Long-Short Temporal Attention network (termed LSTA) is developed for unsupervised VOS task from a holistic view to achieve nearly linear time complexity for the two light modules.
- Transferable Semi-Supervised Semantic Segmentation27
A novel transferable semi-supervised semantic segmentation model that can transfer the learned segmentation knowledge from a few strong categories with pixel- level annotations to unseen weak categories with only image-level annotations is proposed, significantly broadening the applicable territory of deep segmentation models.
- 26
- 19
- Pixelwise Estimation of Signal-Dependent Image Noise Using Deep Residual Learning19
This paper proposes a deep convolutional neural network named “deep residual noise estimator” (DRNE) for pixelwise noise-level estimation and carefully design the architecture of the DRNE, which consists of a stack of customized residual blocks without any pooling or interpolation operation.
- ThermalGS: Dynamic 3D Thermal Reconstruction with Gaussian Splatting18
This work proposes a novel dynamic thermal 3D reconstruction method, named ThermalGS, based on 3D Gaussian Splatting (3DGS), which employs a data-driven approach to directly learn both scene structure and dynamic thermal representation, using RGB and TIR images as input.
- 15
- A photo-responsive self-healing hydrogel loaded with immunoadjuvants and MoS 2 nanosheets for combating post-resection breast cancer recurrence14
An implantable photo-responsive self-healing hydrogel loaded with MoS2 nanosheets and the immunoadjuvant R837 for in situ generation of tumor-associated antigens at the post-surgical site of the primary tumor, enabling sustained and effective activation of the immune response.
- A nanoparticle-containing polycaprolactone implant for combating post-resection breast cancer recurrence12
This strategy using implantable drug-containing polymeric scaffolds to deliver chemo drugs directly to the post-resection site, and simultaneously provide mechanical support and regenerative niche for breast tissue reconstruction holds great promise as an adjuvant localized chemotherapy in clinics for combating post- resection breast cancer recurrence.
- 10
- MSBA: Multiple Scales, Branches and Attention Network With Bag of Tricks for Person Re-Identification10
A model that combines the most effective designs, including multi-scale, multi-branch and attention mechanism, and report training tricks that are no less or even more important in improving person Re-ID performance is proposed.
- 9
- 8
- 7
- Denoising real bursts with squeeze‐and‐excitation residual network6
The proposed deep residual model with squeeze-and-excitation modules for the burst denoising performs significantly better than previous state-of-the-art methods V-BM4D and KPN in removing Gaussian noise and is effective in removing real noise using bursts.
- Fast ℓ 1-minimization algorithm for robust background subtraction6
This paper proposes an approximative ℓ1-minimization algorithm with computationally efficient strategies to achieve real-time performance of sparse model-based background subtraction and demonstrates the higher efficiency and effectiveness of the proposed approach compared with those of other competing methods.
- Two-Stage CNN Model for Joint Demosaicing and Denoising of Burst Bayer Images5
A two-stage convolutional neural network model for joint demosaicing and denoising of burst Bayer images with obvious performance advantages and speed advantages compared with similar approaches is proposed.
- 5
- Salient object detection via robust dictionary representation5
A robust representation based on a discriminative dictionary which consists of non-salient and salient templates, which is compared against 17 state-of-the-art methods on 4 popular datasets by 6 evaluation metrics and demonstrates the competitive performance in terms of qualitative and quantitative results.
- A Noisy Videos Background Subtraction Algorithm Based on Dictionary Learning5
A background subtraction algorithm based on dictionary learning and sparse coding for handling low light conditions and the promising robustness of the proposed approach compared with other competing methods is shown.
- Boosting adversarial robustness via feature refinement, suppression, and alignment3
A novel RSA algorithm to counteract adversarial perturbations while maintaining clean accuracy is proposed, which achieves consistently superior performance in both adversarial robustness and clean accuracy over the state-of-the-art.
- Zoom-GAN: learn to colorize multi-scale targets3
This paper proposes a feature scaling network in this paper called Zoom-GAN to improve the colorizing consistency for small objects and boundary and achieves state-of-the-art on three subjective and objective evaluation metrics.
- 3
- Loss-Driven Channel Pruning of Convolutional Neural Networks3
A loss-driven method to prune redundant channels of CNNs by using Taylor expansion technique regarding to scaling and shifting factors, and prunes those channels by percentile threshold to obtain a compact network with less parameters and FLOPs consumption.
- Dual-Task Integrated Network for Fast Pedestrian Detection in Crowded Scenes3
An fast dual-task pedestrian detector with integrated segmentation context ( DTISC) which predicts pedestrian location as well as its pixel-wise segmentation as well as its pixel-wise segmentation is introduced.
- Are RGB-based salient object detection methods unsuitable for light field data?3
This paper attempts to make the 2D methods available for light field SOD by learning to select the best single image from the 4D tensor by learning to select the best single image from the 4D tensor.
- Enhancement of fish-eye imaging quality based on compressive sensing3
This paper proposes the fish-eye camera spatial resolution first, and the sampling rate of different image blocks is improved according to spatial resolution, and results of the numerical experiments show that the method proposed is effective.
- 3
- A Kalman Filter based Video Denoising Method Using Intensity and Structure Tensor3
Experimental results show that the performance of the proposed method is competitive when compared with state-of-the-art video denoising algorithms based on both peak signal-to-noise-ratio and structural similarity evaluations.
- SCC-Loc: A Unified Semantic Cascade Consensus Framework for UAV Thermal Geo-Localization2
This work proposes SCC-Loc, a unified Semantic-Cascade-Consensus localization framework that establishes a new state-of-the-art, suppressing the mean localization error to 9.37 m and providing a 7.6-fold accuracy improvement within a strict 5-m threshold over the strongest baseline.
- Learning to colorize near-infrared images with limited data2
A novel two-stage framework to recover visible color for NIR images with limited data is proposed and a novel bilateral Res-Unet network is proposed to transfer color features on both sides of the encoder and decoder to improve the semantic correctness of the colored image.
- 2
- A CNN channel pruning low-bit framework using weight quantization with sparse group lasso regularization2
This paper proposes a novel framework for CNNs to simultaneously achieve channel pruning and low-bit quantization by combining weight quantization with Sparse Group Lasso (SGL) regularization by Alternating Direction Method of Multipliers (ADMM).
- 2
- Data Separation of L1-minimization for Real-time Motion Detection2
A data separation algorithm with computationally efficient strategies to achieve real-time performance of sparse model based motion detection using the traditional pursuit algorithms as a pre-process step that converts the iterative optimization into linear addition and multiplication operations.
- 1
- 1
- 1
- 1
- Wiener Loss: A Strong Correlative Loss Applied to Conditional GAN for Color Prediction1
A novel loss function, called Wiener Loss, is proposed to constrain the training of colorization network and this loss is fed back into generator network for training to generate better results.
- A Real-Time Cascaded Video Denoising Algorithm Using Intensity and Structure Tensor1
The proposed cascaded video denoising method is competitive with current state-of-the-art video denoising algorithms on both peak signal-to-noise ratio and structural similarity evaluations, particularly when dealing with large-scale noise.
- –
- –
- –
- MDF-Det: Motion-Aware Decoupling and Scene Filtering for Wide Area Small Moving Target Detection–
MDF-Det establishes a unified coarse-to-fine framework that jointly addresses weak-target preservation, dense-target decoupling, and scene-induced false-alarm suppression in wide-area motion imagery.
- –
- –
- –
- A Simplified Low Rank and Sparse Model for Visual Tracking–
A simplified lowrank and sparse model for visual tracking (LRSVT), which is built upon the particle filter framework, is proposed in this paper and shows that the LRSVT algorithm favorably performs against state-of-the-art tracking methods with regard to accuracy and execution time.
- Data Based Color Constancy–
The proposed unsupervised feature-mining kernel method only depends on the color data without any other information, and achieves comparable performance with other state-of-the-art methods.
- Motion Detection Algorithm for Unmanned Aerial Vehicle Nighttime Surveillance–
A universal dictionary for arbitrary scenes is presented and a robust motion detection algorithm based on sparse representation is proposed, demonstrating the robustness of the proposed approach.
- A Low Illumination Environment Motion Detection Method based on Dictionary Learning–
Experimental results show that compared to the mixture of Gaussian model and frame difference method, the proposed method can reach a better result under extreme low illumination circumstance.
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.