Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks130 from public data
- Deep Learning for Person Re-Identification: A Survey and Outlook2,304
A powerful AGW baseline is designed, achieving state-of-the-art or at least comparable performance on twelve datasets for four different Re-ID tasks, and a new evaluation metric (mINP) is introduced, indicating the cost for finding all the correct matches, which provides an additional criteria to evaluate the Re- ID system for real applications.
- 1,257
- 896
- Salient Object Detection in the Deep Learning Era: An In-Depth Survey810
This paper reviews deep SOD algorithms from different perspectives, including network architecture, level of supervision, learning paradigm, and object-/instance-level detection, and looks into the generalization and difficulty of existing SOD datasets.
- 535
- Dynamic Dual-Attentive Aggregation Learning for Visible-Infrared Person Re-identification522
An intra-modality weighted-part attention module to extract discriminative part-aggregated features, by imposing the domain knowledge on the part relationship mining, and a parameter-free dynamic dual aggregation learning strategy to adaptively integrate the two components in a progressive joint training manner.
- Hi-Net: Hybrid-Fusion Network for Multi-Modal MR Image Synthesis399
A novel Hybrid-fusion Network (Hi-Net) is proposed for multi-modal MR image synthesis, which learns a mapping from multi- modal source images to target images, and effectively exploits the correlations among multiple modalities.
- 329
- 305
- Zero-Shot Video Object Segmentation via Attentive Graph Neural Networks302
This work proposes a novel attentive graph neural network (AGNN) for zero-shot video object segmentation (ZVOS) and extends it to an additional task: image object co-segmentation (IOCS), demonstrating the generalizability of the framework.
- ET-Net: A Generic Edge-aTtention Guidance Network for Medical Image Segmentation276
A generic medical segmentation method, called Edge-aTtention guidance Network (ET-Net), which embeds edge-attention representations to guide the segmentation network, which outperforms current state-of-the-art segmentation methods.
- Evaluation of Retinal Image Quality Assessment Networks in Different Color-Spaces251
A simple yet efficient deep network is proposed, named Multiple Color-space Fusion Network (MCF-Net), which integrates the different color-space representations at both a feature-level and prediction-level to predict image quality grades and demonstrates that the performances of automated diagnostic systems are highly dependent on image quality.
- 236
- 235
- 218
- 216
- 214
- MATNet: Motion-Attentive Transition Network for Zero-Shot Video Object Segmentation212
A novel end-to-end learning neural network, i.e., MATNet, for zero-shot video object segmentation (ZVOS), motivated by the human visual attention behavior, which leverages motion cues as a bottom-up signal to guide the perception of object appearance.
- Striking the Right Balance With Uncertainty199
This paper demonstrates that the Bayesian uncertainty estimates directly correlate with the rarity of classes and the difficulty level of individual samples, and presents a novel framework for uncertainty based class imbalance learning that efficiently utilizes sample and class uncertainty information to learn robust features and more generalizable classifiers.
- Face Forensics in the Wild195
This work constructs a novel large-scale dataset, called FFIW10K, which comprises 10,000 high-quality forgery videos, with an average of three human faces in each frame, and proposes a novel algorithm to tackle the task of multi-person face forgery detection.
- 167
- Quadruplet Network with One-Shot Learning for Visual Tracking.162
A new quadruplet deep network is proposed to examine the potential connections among the training instances, aiming to achieve a more powerful representation and is evaluated by model-free tracking-by-detection of objects from a single initial exemplar in several visual object tracking benchmarks.
- Adversarial Defense by Restricting the Hidden Space of Deep Neural Networks161
This work proposes to class-wise disentangle the intermediate feature representations of deep networks, and forces the features for each class to lie inside a convex polytope that is maximally separated from the polytopes of other classes.
- LiDAR-Based Online 3D Video Object Detection With Graph-Based Message Passing and Spatiotemporal Transformer Attention158
An end-to-end online 3D video object detector that operates on point cloud sequences that enhances the conventional ConvGRU with an attentive memory gating mechanism is proposed.
- 157
- 156
- Distilled Siamese Networks for Visual Tracking148
This model is intuitively inspired by the one teacher versus multiple students learning method typically employed in schools and contains a single teacher-student distillation module and a student-student knowledge sharing mechanism that is designed using a tracking-specific distillation strategy to transfer knowledge from a teacher to students.
- 142
- 139
- Cascaded Human-Object Interaction Recognition130
This work introduces a cascade architecture for a multi-stage, coarse-to-fine HOI understanding, and makes the framework flexible to perform fine-grained pixel-wise relation segmentation; this provides a new glimpse into better relation modeling.
- 130
- 130
- Paying Attention to Video Object Pattern Understanding128
For the first time, it is quantitatively verified the high consistency of visual attention behavior among human observers, and found strong correlation between human attention and explicit primary object judgments during dynamic, task-driven viewing.
- Weakly Supervised 3D Object Detection from Lidar Point Cloud127
This work proposes a weakly supervised approach for 3D object detection, only requiring a small set of weakly annotated scenes, associated with a few precisely labeled object instances, achieved by a two-stage architecture design.
- 119
- 114
- Tree Energy Loss: Towards Sparsely Annotated Semantic Segmentation109
A novel tree energy loss for SASS is proposed by providing semantic guidance for unlabeled pixels by sequentially applying these affinities to the net-work prediction, achieving dynamic online self-training.
- 107
- 96
- Graph Neural Network and Spatiotemporal Transformer Attention for 3D Video Object Detection From Point Clouds94
This paper proposes an Attentive Spatiotemporal Transformer GRU (AST-GRU), which contains a Spatial Transformer Attention (STA) module and a Temporal Trans transformer Attention (TTA) module that enhances the vanilla GRU to focus on small objects and better align the moving objects.
- Thread of Thought Unraveling Chaotic Contexts93
This work introduces the "Thread of Thought" strategy, which draws inspiration from human cognitive processes and systematically segments and analyzes extended contexts while adeptly selecting pertinent information within the chaotic context.
- Reducing Estimation Bias via Triplet-Average Deep Deterministic Policy Gradient91
This article investigates the underestimation phenomenon in the recent twin delay deep deterministic actor-critic algorithm and theoretically demonstrates its existence, and proposes a novel triplet-averageDeep deterministic policy gradient algorithm that takes the weighted action value of three target critics to reduce the estimation bias.
- 88
- 87
- 86
- 84
- 83
- DI-V2X: Learning Domain-Invariant Representation for Vehicle-Infrastructure Collaborative 3D Object Detection74
DI-V2X is proposed, that aims to learn Domain-Invariant representations through a new distillation framework to mitigate the domain discrepancy in the context of V2X 3D object detection.
- 71
- Augmented reality based real-time subcutaneous vein imaging system62
A novel 3D reconstruction and fast imaging system for subcutaneous veins by augmented reality to reduce the failure rate and time required in intravenous injection by providing augmented vein structures that back-project superimposed veins on the skin surface of the hand.
- 60
- 59
- 54
- Adaptive Siamese Tracking with a Compact Latent Network52
This article provides an intuitive viewing to simplify the Siamese-based trackers by converting the tracking task to a classification, and presents a statistics-based compact latent feature for fast adjustment by efficiently extracting the sequence-specific information.
- 52
- NETNet: Neighbor Erasing and Transferring Network for Better Single Shot Object Detection45
A new Neighbor Erasing and Transferring (NET) mechanism to reconfigure the pyramid features and explore scale-aware features is proposed, which achieves a better trade-off for real-time and accurate object detection.
- 45
- 43
- DrivingSphere: Building a High-fidelity 4D World for Closed-loop Simulation39
By providing a dynamic and realistic simulation environment, DrivingSphere enables comprehensive testing and validation of autonomous driving algorithms, ultimately advancing the development of more reliable autonomous cars.
- Rethinking Visual Dependency in Long-Context Reasoning for Large Vision-Language Models39
This study empirically analyze LVLMs in long-context reasoning, revealing that increased context length leads to a higher dependence on language at the expense of visual dependency, and proposes a novel training-free context pruning method that selectively removes less critical textual information.
- 39
- 36
- Self-Learning With Rectification Strategy for Human Parsing36
This paper proposes a trainable graph reasoning method that establishes internal structural connections between graph nodes to correct two typical errors in the pseudo-labels, i.e., the global structural error and the local consistency error.
- Weakly Supervised Monocular 3D Object Detection Using Multi-View Projection and Direction Consistency32
This work proposes a new weakly supervised monocular 3D objection detection method, which can train the model with only 2D labels marked on images, and proposes a weakly-supervised architecture based on these consistencies.
- 32
- Bilateral Cross-Modality Graph Matching Attention for Feature Fusion in Visual Question Answering32
A graph matching attention (GMA) network that not only buildsgraph for the image but also constructs graph for the question in terms of both syntactic and embedding information and achieves the state-of-the-art performance on the GQA dataset and the VQA 2.0 dataset.
- Deep Multi-modal Latent Representation Learning for Automated Dementia Diagnosis32
A novel Deep Latent Multi-modality Dementia Diagnosis framework based on a deep non-negative matrix factorization (NMF) model that integrates the feature fusion/learning process into the classifier construction step for eliminating the gap between neuroimaging features and disease labels is proposed.
- 31
- Consistency and Diversity Induced Human Motion Segmentation29
A novel Consistency and Diversity induced human Motion Segmentation (CDMS) algorithm that factorizes the source and target data into distinct multi-layer feature spaces, in which transfer subspace learning is conducted on different layers to capture multi-level information.
- Is a 3D-Tokenized LLM the Key to Reliable Autonomous Driving?26
This work introduces DETR-style 3D perceptrons as 3D tokenizers, which connect LLM with a one-layer linear projector and demonstrates superior performance in both 3D detection and ego planning tasks on nuScenes dataset, proving that 3D-tokenized LLM is the key to reliable autonomous driving.
- 26
- 26
- $$\text {M}^2\text {Net}$$: Multi-modal Multi-channel Network for Overall Survival Time Prediction of Brain Tumor Patients24
An end-to-end OS time prediction model, Multi-modal Multi-channel Network, which first project the 3D MR volume onto 2D images in different directions, which reduces computational costs, while preserving important information and enabling pre-trained models to be transferred from other tasks.
- Robust geometric ℓ p -norm feature pooling for image classification and action recognition24
This paper proposes to generalize previous pooling methods toward a weighted źp-norm spatial pooling function tailored for class-specific feature spatial distribution, and proposes a simple yet effective self-alignment step during both learning and testing to adaptively adjust the pooling weights for individual images.
- RAGNet: Large-Scale Reasoning-Based Affordance Segmentation Benchmark Towards General Grasping22
A large-scale grasping-oriented affordance segmentation benchmark with human-like instructions, named RAGNet, and a comprehensive affordance-based grasping framework, named AffordanceNet, which consists of a VLM pretrained on the authors' massive affordance data and a grasping network that conditions an affordance map to grasp the target.
- Clip Fusion with Bi-level Optimization for Human Mesh Reconstruction from Monocular Videos21
A bi-level clip inference network for HMR is presented, which leverages both local motion and global context explicitly for dense 3D reconstruction and investigates an alternative perspective by treating video-based HMR as clip-wise inference.
- Teacher-Students Knowledge Distillation for Siamese Trackers21
This paper proposes a novel distilled Siamese tracker framework to learn small, fast yet accurate trackers (students), which can capture critical knowledge from large SiamesE tracker (teachers) by a teacher-students knowledge distillation model.
- 20
- 20
- 19
- MAGIC CUBE TRANSFORMATION AND ITS APPLICATION IN DIGITAL IMAGE ENCRYPTION18
The results indicate that the proposed algorithm based on Magic Cube Transformation can get satisfying effect and the characteristics of the algorithm are summarized and the aspects of the subsequent work are prospected.
- 17
- 16
- 16
- Inter-modality Dependence Induced Data Recovery for MCI Conversion Prediction16
This work proposes a novel diagnostic model that integrates missing data recovery, latent space learning and prediction model learning into a unified framework, and shows the effectiveness of this method on the Alzheimer’s Disease Neuroimaging Initiative dataset.
- 15
- DC-ControlNet: Decoupling Inter- and Intra-Element Conditions in Image Generation with Diffusion Models14
Extensive evaluations show that DCControlNet significantly outperforms existing ControlNet models and Layout-to-Image generative models in terms of control flexibility and precision in multi-condition control.
- Semantic Causality-Aware Vision-Based 3D Occupancy Prediction14
This paper designs a novel causal loss that enables holistic, end-to-end supervision of the modular 2D-to-3D transformation pipeline, and proposes the Semantic Causality-Aware 2D-to-3D Transformation, which comprises three components guided by the causal loss.
- Rethinking Temporal Fusion with a Unified Gradient Descent View for 3D Semantic Occupancy Prediction14
GDFusion opens up the underexplored aspects of temporal fusion within the VisionOcc framework, focusing on both temporal cues and fusion strategies, and proposes a novel fusion strategy by reinterpreting the formulation of vanilla RNNs.
- 14
- 14
- 12
- 12
- Decoupling Fine Detail and Global Geometry for Compressed Depth Map Super-Resolution11
A novel framework for compressed depth map super-resolution that decouples the high-quality depth map reconstruction process by handling global and detailed geometric features separately is proposed, and significantly outperforms current methods in terms of geometric consistency and detail recovery.
- Towards Better Cephalometric Landmark Detection With Diffusion Data Generation11
This work has developed an innovative data generation method capable of producing diverse cephalometric X-ray images along with corresponding annotations without human intervention and introduces large-scale vision detection models into the cephalometric landmark detection task to improve accuracy.
- Condition Errors Refinement in Autoregressive Image Generation with Diffusion Loss10
A theoretical comparison of conditional diffusion and autoregressive diffusion with diffusion loss is presented, demonstrating that patch denoising optimization in autoregressive models effectively mitigates condition errors and leads to a stable condition distribution.
- 10
- 10
- 9
- 8
- 8
- 8
- TextFormer: A Query-based End-to-end Text Spotter with Mixed Supervision7
TextFormer is proposed, a query-based end-to-end text spotter with a transformer architecture that overcomes the suboptimization problem of RoI operations and design an adaptive global aggregation (AGG) module to transfer global features into sequential features for reading arbitrarily-shaped texts.
- Understanding and Correcting Low-quality Retinal Fundus Images for Clinical Analysis.7
A clinical-oriented fundus enhancement network is proposed to suppress the global degradation factors, and simultaneously preserve anatomical retinal structures and pathological characteristics for clinical observation and analysis, and it is shown that the fundus correction method can benefit medical image analysis applications, e.g, retinal vessel segmentation and optic disc/cup detection.
- Automatic Schelling Point Detection From Meshes6
The experimental results demonstrate that DS-Net is capable of detecting mesh Schelling points effectively and outperforms various state-of-the-art mesh saliency methods and deep learning models, both qualitatively and quantitatively.
- 6
- 6
- HiCoGen: Hierarchical Compositional Text-to-Image Generation in Diffusion Models via Reinforcement Learning4
This work proposes a Hierarchical Compositional Generative framework (HiCoGen) built upon a novel Chain of Synthesis (CoS) paradigm, and introduces a reinforcement learning (RL) framework that significantly outperforms existing methods in both concept coverage and compositional accuracy.
- 3
- 3
- RAWMamba: Unified sRGB-to-RAW De-rendering With State Space Model3
The core of RAWMamba is the Unified Metadata Embedding module, which harmonizes diverse metadata types into a unified representation, and a multi-perspective affinity modeling method is proposed to promote the extraction of reference information.
- 3
- 3
- Accelerating Training of Autoregressive Video Generation Models via Local Optimization with Representation Continuity2
A Local Optimization (Local Opt.) method, which optimizes tokens within localized windows while leveraging contextual information to reduce error propagation, is proposed, which achieves superior performance to the baseline while halving the training cost without sacrificing quality.
- Multimodal Large Language Models for Multi-Subject In-Context Image Generation2
This work proposes MUSIC, the first MLLM specifically designed for multi-subject in-context generation and introduces an automatic and scalable data generation pipeline that eliminates the need for manual annotation, and curates MSIC, a new benchmark tailored for multi-subject in-context generation.
- 2
- A hierarchical visual saliency detection method by combining distinction and background probability maps2
A bottom-up method for image saliency detection by combining the distinction map and the background probability map is proposed and incorporated into a hierarchical framework to generate the final saliency map.
- 1
- –
- –
- –
- –
- –
- TransBridge: Boost 3D Object Detection by Scene-Level Completion With Transformer Decoder–
This paper proposes TransBridge, a novel transformer-based up-sampling block that fuses the features from the detection and completion networks and employs the transformer mechanism to establish connections between channels and spatial relations, resulting in a high-resolution feature map used for completion purposes.
- –
- CubeFormer: A Simple yet Effective Baseline for Lightweight Image Super-Resolution–
This work proposes a simple yet effective baseline called CubeFormer, designed to enhance feature richness by completing holistic information aggregation, and introduces cube attention, which expands 2D attention to 3D space, facilitating exhaustive information interactions, further encouraging comprehensive information extraction and promoting feature variety.
- Selective Video Cutout using Global Pyramid Models and Local Uncertainty Propagation.–
A pyramid histogram based confidence map that incorporates structure information into appearance statistics that combines geodesic distance based dynamic models and employs an efficient measure of uncertainty propagation using local classifiers to determine the image regions where the object labels might be ambiguous.
- –
- –
- An image-segment algorithm based on prewitt method and its applications–
An improved image-segment method of Prewitt was proposed based on the technology of histogram equalization and experimental results indicated that the algorithm proposed here could obtain better outcome of segmentation than the P Drewitt method.
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.