Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks263 from public data
- Uformer: A General U-Shaped Transformer for Image Restoration2,405
Uformer, an effective and efficient Transformer-based architecture for image restoration, in which a hierarchical encoder-decoder network is built using the Transformer block and a learnable multi-scale restoration modulator in the form of a multi- scale spatial bias to adjust features in multiple layers of the Uformer decoder is proposed.
- 1,104
- Neighbor2Neighbor: Self-Supervised Denoising from Single Noisy Images489
The proposed Neighbor2Neighbor framework is able to enjoy the progress of state-of-the-art supervised denoising networks in network architecture design and avoids heavy dependence on the assumption of the noise distribution.
- Gabor Convolutional Networks420
A new deep model is proposed, termed Gabor Convolutional Networks (GCNs or Gabor CNNs), which incorporates Gabor filters into DCNNs to enhance the resistance of deep learned features to the orientation and scale changes.
- Prompt Distribution Learning367
This work presents prompt distribution learning for effectively adapting a pre-trained vision-language model to address downstream recognition tasks and employs a Gaussian distribution to model them effectively and derive a surrogate loss for efficient training.
- VastGaussian: Vast 3D Gaussians for Large Scene Reconstruction349
This work proposes a progressive partitioning strategy to divide a large scene into multiple cells, where the training cameras and point cloud are properly distributed with an airspace-aware visibility criterion, and introduces decoupled appearance modeling into the optimization process to reduce appearance variations in the rendered images.
- Diffusion Model-Based Image Editing: A Survey328
This survey provides an exhaustive overview of existing methods using diffusion models for image editing, covering both theoretical and practical aspects in the field, and explores both earlier traditional context-driven and current multimodal conditional methods.
- 276
- 263
- 206
- 206
- 192
- 187
- Domain General Face Forgery Detection by Learning to Weight179
This paper argues that different faces contribute differently to a detection model trained on multiple domains, making the model likely to fit domain-specific biases, and proposes the LTW approach based on the meta-weight learning algorithm, which configures different weights for face images from different domains.
- Multiple Instance Active Learning for Object Detection170
MI-AOD defines an instance uncertainty learning module, which leverages the discrepancy of two adversarial instance classifiers trained on the labeled set to predict instance uncertainty of the unlabeled set, and estimates the image uncertainty by re-weighting instances in a multiple instance learning (MIL) fashion.
- WaveDM: Wavelet-Based Diffusion Models for Image Restoration160
Evaluations on twelve benchmark datasets demonstrate that WaveDM achieves state-of-the-art performance with the efficiency that is comparable to traditional one-pass methods and over 100× faster than existing image restoration methods using vanilla diffusion models.
- Wavelet-Based Dual-Branch Network for Image Demoiréing158
Although designed for image demoireing, WDNet has been applied to two other low-levelvision tasks, outperforming state-of-the-art image deraining and derain-drop methods on the Rain100h and Raindrop800 data sets, respectively.
- 152
- SketchyCOCO: Image Generation From Freehand Scene Sketches148
This work introduces the first method for automatic image generation from scene-level freehand sketches that allows for controllable image generation by specifying the synthesis goal via free hand sketches and builds a large-scale composite dataset called SketchyCOCO to support and evaluate the solution.
- Exploiting Kernel Sparsity and Entropy for Interpretable CNN Compression142
KSE is capable of simultaneously compressing each layer in an efficient way, which is significantly faster compared to previous data-driven feature map pruning methods, and significantly outperforms state-of-the-art methods.
- A comparison of 3D shape retrieval methods based on a large-scale benchmark supporting multimodal queries142
A more comprehensive comparison of twenty-six 3D shape retrieval methods is performed by evaluating them on the common benchmark, compiled to be a superset of existing benchmarks and presents a new challenge to retrieval methods.
- A spatial-temporal approach for video caption detection and recognition141
A video caption detection and recognition system based on a fuzzy-clustering neural network (FCNN) classifier that improves the Chinese video-caption recognition accuracy from 13% to 86% on a set of news video captions.
- Pairwise constraint propagation by semidefinite programming for semi-supervised classification137
The general problem of learning from both pairwise constraints and unlabeled data is considered and it is proposed to learn a mapping that is smooth over the data graph and maps the data onto a unit hypersphere, where two must-link objects are mapped to the same point while two cannot- link objects are maps to be orthogonal.
- 134
- 133
- Geometric Reinforcement Learning for Path Planning of UAVs132
Extensive experimental results validate the effectiveness and feasibility of GRL on the navigation of UAVs.
- 125
- Automatic facial expression recognition on a single 3D face by exploring shape deformation122
Unlike previous methods which recognize a facial expression with the help of manually labeled key points and/or a neutral face, this method works on a single 3D face without any manual assistance.
- Motion Deblurring with Real Events120
An end-to-end learning framework for event-based motion deblurring in a self-supervised manner, where real-world events are exploited to alleviate the performance degradation caused by data inconsistency, and a piecewise linear motion model is proposed to take into account motion non-linearities.
- 117
- Multinomial Distribution Learning for Effective Neural Architecture Search114
A Multinomial Distribution Learning for extremely effective NAS is proposed, which considers the search space as a joint multinomial distribution, and the optimal network structure is obtained by the operations with the most likely probability in this distribution.
- A maximum entropy feature descriptor for age invariant face recognition110
A new maximum entropy feature descriptor (MEFD) is developed that encodes the microstructure of facial images into a set of discrete codes in terms of maximum entropy, to overcome the representation and matching problems in age invariant face recognition.
- ReCU: Reviving the Dead Weights in Binary Neural Networks109
It is proved that reviving the "dead weights" by ReCU can result in a smaller quantization error, and also takes into account the information entropy of the weights, and mathematically analyze why the weight standardization can benefit BNNs.
- Filter Sketch for Network Pruning106
FilterSketch encodes the second-order information of pretrained weights, which enables the representation capacity of pruned networks to be recovered with a simple fine-tuning procedure, leading to a several-orders-of-magnitude reduction of time cost in the optimization of pruning.
- Multiple Anchor Learning for Visual Object Detection106
A Multiple Instance Learning (MIL) approach that selects anchors and jointly optimizes the two modules of a CNN-based object detector, referred to as Multiple Anchor Learning (MAL), which improves the baseline RetinaNet with significant margins on the commonly used MS-COCO object detection benchmark and achieves new state-of-the-art detection performance compared with recent methods.
- Towards Compact CNNs via Collaborative Compression105
A Collaborative Compression scheme, which joints channel pruning and tensor decomposition to compress CNN models by simultaneously learning the model sparsity and low-rankness and proposes multi-step heuristic compression to remove redundant compression units step-by-step.
- IntraQ: Learning Synthetic Images with Intra-Class Heterogeneity for Zero-Shot Network Quantization102
This paper proposes a novel zero-shot quantization method called IntraQ, which employs a local object reinforcement that locates the target objects at different scales and positions of the synthetic images and introduces a marginal distance constraint to form class-related features distributed in a coarse area.
- Multi-Target Domain Adaptation with Collaborative Consistency Learning101
The proposed collaborative learning framework can effectively exploit rich structured information contained in both labeled source domain and multiple unlabeled target domains and performs favorably against state-of-the-art unsupervised domain adaptation methods specially trained on a single source-target pair.
- Diversity Matters: Fully Exploiting Depth Clues for Reliable Monocular 3D Object Detection100
A depth solving system that fully explores the visual clues from the subtasks in M3OD and generates multiple estimations for the depth of each target, and develops a depth selection and combination strategy that removes abnormal estimations caused by collapsed assumptions and adaptively combines the remaining estimations into a single one.
- Projection Convolutional Neural Networks for 1-bit CNNs via Discrete Back Propagation100
This paper introduces projection convolutional neural networks with a discrete back propagation via projection (DBPP) to improve the performance of binarized neural networks (BNNs) and learns a set of diverse quantized kernels that compress the full-precision kernels in a more efficient way than those proposed previously.
- 99
- Multi-Source Domain Adaptation with Collaborative Learning for Semantic Segmentation97
This paper proposes a novel multi-source domain adaptation framework based on collaborative learning for semantic segmentation that significantly outperforms all previous state-of-the-arts single-source and multi- source unsupervised domain adaptation methods.
- Graph-based method for face identification from a single 2D line drawing95
A graph-based optimization method is proposed for identifying the faces is a line drawing that generates the same results of face identification as Shpitalni and Lipson's method, but is much faster when dealing with objects of more than 20 faces.
- CoSeR: Bridging Image and Language for Cognitive Super-Resolution93
The Cognitive Super-Resolution (CoSeR) framework is introduced, empowering SR models with the capacity to comprehend low-resolution images and a novel condition injection scheme called “Ali-in-Attention ”, consolidating all conditional information into a single module.
- Context-Transformer: Tackling Object Confusion for Few-Shot Detection93
A novel Context-Transformer is proposed within a concise deep transfer framework that can effectively leverage source-domain object knowledge as guidance, and automatically exploit contexts from only a few training images in the target domain to reduce object confusion in few-shot scenarios.
- 92
- Low-Light Image Enhancement with Illumination-Aware Gamma Correction and Complete Image Modelling Network85
A novel Transformer block is proposed to completely simulate the dependencies of all pixels across images via a local-to-global hierarchical attention mechanism, so that dark areas could be inferred by borrowing the information from far informative regions in a highly effective manner.
- ZONE: Zero-Shot Instruction-Guided Local Editing84
This work proposes a Zero-shot instructiON-guided local image Editing approach, termed ZONE, which allows for arbitrary manipulation of a specific region with a single instruction while preserving the rest.
- Circulant Binary Convolutional Networks: Enhancing the Performance of 1-Bit DCNNs With Circulant Back Propagation84
Newcirculant filters (CiFs) and a circulant binary convolution (CBConv) are proposed to enhance the capacity of binarized convolutional features via the authors' circulants back propagation (CBP).
- 84
- Visual object tracking via sample-based Adaptive Sparse Representation (AdaSR)83
A new approach for visual object tracking based on Sample-Based Adaptive Sparse Representation (AdaSR), which ensures that the tracked object is adaptively and compactly expressed with predefined samples, which is better than those of several representative tracking methods.
- Cooperative and Geometric Learning Algorithm (CGLA) for path planning of UAVs with limited information81
Risk evaluation is introduced first time in this paper for UAV navigation and extensive computer simulation results validate the effectiveness and feasibility of CGLA for safe navigation of multiple UAVs.
- AttriCLIP: A Non-Incremental Learner for Incremental Knowledge Learning78
This paper proposes a non-incremental learner, named AttriCLIP, to incrementally extract knowledge of new classes or tasks, built upon the pre-trained visual-language model CLIP and compares it with CLIP-based and previous state-of-the-art continual learning methods in realistic settings with domain-shift and long-sequence learning.
- 77
- Semi-supervised Domain Adaptation based on Dual-level Domain Mixing for Semantic Segmentation77
This work focuses on a more practical setting of semi-supervised domain adaptation (SSDA) where both a small set of labeled target data and large amounts of labeled source data are available and a novel framework based on dual-level domain mixing is proposed.
- 75
- ADAPT: Vision-Language Navigation with Modality-Aligned Action Prompts74
This paper proposes modAlity-aligneD Action PrompTs (ADAPT), which provides the VLN agent with action prompts to enable the explicit learning of action-level modality alignment to pursue successful navigation and shows the superiority of ADAPT over state-of-the-art methods.
- 73
- 72
- 70
- 66
- Uncertainty-Driven Dehazing Network65
A novel uncertainty-driven dehazing network (UDN) is proposed that improves the dehazed results by exploiting the relationship between the uncertain and confident representations, and develops an uncertainty- driven self-distillation loss to improve the uncertain representation by transferring the knowledge from the confident one.
- ViewCo: Discovering Text-Supervised Segmentation Masks via Multi-View Semantic Consistency62
The text-to-views consistency modeling benefits the dense assignment of the visual features by encouraging different crops to align with the same text, while the cross-view segmentation consistency modeling provides additional self-supervision, overcoming the limitation of ambiguous text supervision for segmentation masks.
- Face Animation with an Attribute-Guided Diffusion Model61
An Attribute-Guided Conditioning Network (AGCN) is designed to adaptively combine the coarse animation features and 3D face reconstruction results, which can incorporate appearance and motion conditions into the diffusion process to mitigate the uncontrollable synthesis effect of the diffusion model.
- Semi-supervised sparse metric learning using alternating linearization optimization61
Compared with conventional methods, this approach can effectively take advantage of semi-supervision and automatically discover the sparse metric structure underlying input data patterns, and demonstrates the efficacy of the proposed approach with extensive experiments carried out on six datasets.
- 61
- Identifying faces in a 2D line drawing representing a manifold object60
A number of properties, most of which relate to planar manifold geometry in line drawings, are presented to identify most of the cycles that are or are not real faces in a drawing, thus reducing the number of unknown cycles in the second searching.
- Bayesian Optimized 1-Bit CNNs58
This paper proposes a novel approach, called Bayesian optimized 1-bit CNNs (denoted as BONNs), taking the advantage of Bayesian learning, a well-established strategy for hard problems, to significantly improve the performance of extreme 1- bit CNNs.
- 58
- Unsupervised Image Super-Resolution with an Indirect Supervised Path57
A novel framework which is composed of two stages: unsupervised image translation between real LR and synthetic LR images; and supervised super-resolution from approximated real LR images to the paired HR images, which achieves very good performance on datasets of NTIRE 2017, NTIRE 2018 and NTIRE 2020.
- One-two-one networks for compression artifacts reduction in remote sensing57
An in-depth investigation into the OTO architecture based on the Taylor expansion is provided, which shows that these two kinds of information can be fused in a nonlinear scheme to gain more capacity of handling complicated image compression artifacts, especially the blocking effect in compression.
- Projection & Probability-Driven Black-Box Attack54
This paper proposes Projection & Probability-driven Black-box Attack (PPBA), a method to tackle the problem of generating adversarial examples in a black-box setting by reducing the solution space and providing better optimization.
- An End-to-End Foreground-Aware Network for Person Re-Identification53
This work proposes an end-to-end foreground-aware network to discriminate against the foreground from the background by learning a soft mask for person re-identification and demonstrates the effectiveness of this approach on three challenging datasets.
- Dual Distribution Alignment Network for Generalizable Person Re-Identification53
This paper proposes an end-to-end Dual Distribution Alignment Network (DDAN) to learn domain-invariant features with dual-level constraints: the domain-wise adversarial feature learning and the identity-wise similarity enhancement.
- 53
- Bayesian tensor inference for sketch-based facial photo hallucination53
This paper develops a statistical inference approach, Bayesian Tensor Inference, for style transformation between photo images and sketch images of human faces, motivated by the rationale that image appearance is determined by two cooperative factors: image content and image style.
- 51
- 49
- CoupAlign: Coupling Word-Pixel with Sentence-Mask Alignments for Referring Image Segmentation48
CoupAlign is proposed, a simple yet effective multi-level visual-semantic alignment method to couple sentence-mask alignment with word-pixel alignment to enforce object mask constraint for achieving more accurate localization and segmentation.
- High-Order Residual Network for Light Field Super-Resolution48
Experimental results show that the proposed high-order residual network enables high-quality reconstruction even in challenging regions and outperforms state-of-the-art single image or LF reconstruction methods with both quantitative measurements and visual evaluation.
- 48
- Image inpainting by global structure and texture propagation48
A novel global algorithm for region filling is proposed for image inpainting where the filling process is formulated as an energy minimization problem by Markov random fields and the belief propagation is utilized to solve the problem.
- Evolutionary Search for Faces from Line Drawings48
A new method to tackle the face identification problem by a variable-length genetic algorithm with a novel heuristic and geometric constraints incorporated for local search is proposed.
- 46
- 45
- 44
- 43
- Large-Scale Few-Shot Learning via Multi-modal Knowledge Discovery43
A method based on multi-modal knowledge discovery that uses the visual knowledge to help the feature extractors focus on different visual parts and develops three schemes to minimize the prediction error and balance the training procedure.
- 42
- 41
- Multi-feature canonical correlation analysis for face photo-sketch image retrieval40
The MCCA is an extension and improvement of the canonical correlation analysis (CCA) algorithm using multiple features combined with two different random sampling methods in feature space and sample space to improve the performance of automatic face photo-sketch image retrieval.
- 40
- 39
- 39
- Generalizing Event-Based Motion Deblurring in Real-World Scenarios38
This work proposes a scale-aware network that allows flexible input spatial scales and enables learning from different temporal scales of motion blur, and develops a two-stage self-supervised learning scheme to fit real-world data distribution.
- 38
- 38
- 38
- 37
- 37
- Feature Calibration Network for Occluded Pedestrian Detection34
A novel feature learning method in the deep learning framework, referred to as Feature Calibration Network (FC-Net), to adaptively detect pedestrians under various occlusions, which improves detection performance on occluded pedestrians up to 10% while maintaining excellent performance on non-occluded instances.
- Correctable Landmark Discovery via Large Models for Vision-Language Navigation33
In CONSOLE, VLN is cast as an open-world sequential landmark discovery problem, by introducing a novel correctable landmark discovery scheme based on two large models ChatGPT and CLIP, and a learnable cooccurrence scoring module, which corrects the importance of each cooccurrence according to actual observations for accurate landmark discovery.
- GPT4Motion: Scripting Physical Motions in Text-to-Video Generation via Blender-Oriented GPT Planning30
- AsConvSR: Fast and Lightweight Super-Resolution Network with Assembled Convolutions30
This work analyzes the applications of the idea of divide-and-conquer in super-resolution, and proposes assembled convolutions which can adapt convolution kernels according to the input features, and outperforms all the state-of-the-art efficient super- resolution models.
- 30
- 28
- 28
- 27
- 27
- 27
- 26
- 26
- 26
- 26
- MixReorg: Cross-Modal Mixed Patch Reorganization is a Good Mask Learner for Open-World Semantic Segmentation25
MixReorg is proposed, a novel and straightforward pre-training paradigm for semantic segmentation that enhances a model’s ability to reorganize patches mixed across images, exploring both local visual relevance and global semantic coherence.
- 25
- 24
- 23
- 23
- 23
- 23
- 22
- 22
- 22
- Towards Medical Artificial General Intelligence via Knowledge-Enhanced Multimodal Pretraining21
The proposed MOTOR successfully mimics the human practice of fulfilling a"medical student" to accelerate the process of becoming a"specialist" and believes that this work makes a significant stride in realizing MAGI.
- 21
- 21
- 20
- 20
- Wavelet-based network for high dynamic range imaging19
A novel frequency-guided end-to-end deep neural network (FHDRNet) is proposed to conduct HDR fusion in the frequency domain, and Discrete Wavelet Transform (DWT) is used to decompose inputs into different frequency bands.
- 19
- 18
- 18
- Language-Driven Visual Consensus for Zero-Shot Semantic Segmentation18
Equipped with a vision-language prompting strategy, this approach significantly boosts the generalization capacity of segmentation models for unseen classes and introduces route attention into the transformer decoder to find visual consensus, thereby enhancing semantic consistency within the same object.
- 18
- 18
- 17
- 16
- Sketch-based 3-D modeling for piecewise planar objects in single images16
Experimental results on various objects show that the proposed approach provides a good solution to these three problems: requiring rough sketch input only, accurate modeling for a large class of objects, and automatically recovering the invisible part of an object and providing a complete 3-D model.
- 16
- 16
- 15
- 15
- 15
- 15
- 15
- 15
- 15
- 14
- 14
- 14
- 13
- 13
- 13
- 13
- 13
- 12
- 12
- 12
- 12
- RelCLIP: Adapting Language-Image Pretraining for Visual Relationship Detection via Relational Contrastive Learning11
A novel Relational Contrastive Learning (RCL) approach which explores relation-level visual-semantic alignment via learning to match cross-modal relational embeddings, and achieves significant improvements over strong baselines under full and distant supervision, demonstrating its powerful generalization ability in learning relationship representations.
- 11
- 11
- 11
- 11
- 11
- Learning Enriched Illuminants for Cross and Single Sensor Color Constancy10
A cross-sensor self-supervised training to train the network that outperform other state-of-the-art methods on cross and single sensor evaluations, respectively, with only 16% parameters of the previous best model.
- 10
- 10
- 10
- 10
- 9
- API-Net: Robust Generative Classifier via a Single Discriminator9
An Anti-Perturbation Inference (API) method, which searches for anti-perturbations to maximize the lower bound of the joint log-likelihood of inputs and classes, and is constructed upon a generative classifiers upon a single discriminator, which takes advantage of the generative properties to tackle off-manifold examples while maintaining a succinct structure for effective optimization.
- 8
- 8
- 8
- 8
- 8
- 8
- 8
- 7
- 7
- 7
- 7
- 6
- 6
- 6
- 6
- 6
- 6
- 5
- 5
- 4
- 4
- 4
- 4
- 4
- 4
- 4
- 4
- An example-based approach to 3D man-made object reconstruction from line drawings4
An example-based approach to reconstructing 3D object, either planar or curved, from a single-view line drawing with sketch errors, which can handle a wide range of 3D man-made objects including curved components.
- 4
- 4
- 3
- 3
- 3
- 3
- Learning Unorthogonalized Matrices for Rotation Estimation3
By replacing the orthogonalization incorporated representation with the proposed PRoM in various rotation-related tasks, this work achieves state-of-the-art results on large-scale benchmarks for human pose estimation.
- 3
- 3
- 3
- 3
- 3
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
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 2026-10-13. 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.