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
- Cheng‐Lin LiuSuggested from co-authorship
Is this you? Claim this profile to confirm or dismiss it.
Works116 from public data
- 312
- 229
- Non-Salient Region Object Mining for Weakly Supervised Semantic Segmentation223
A graph-based global reasoning unit is introduced to strengthen the classification network’s ability to capture global relations among disjoint and distant regions and a potential object mining module is proposed to reduce the false-negative rate in pseudo labels.
- TransZero: Attribute-Guided Transformer for Zero-Shot Learning206
This paper proposes an attribute-guided Transformer network to learn the attribute localization for discriminative visual-semantic embedding representations in ZSL, termed TransZero, which achieves a new state-of-the-art on three ZSL benchmarks.
- Unsupervised Domain Adaptation with Noise Resistible Mutual-Training for Person Re-identification186
A Noise Resistible Mutual-Training (NRMT) method, which maintains two networks during training to perform collaborative clustering and mutual instance selection, which outperforms the state-of-the-art UDA methods for person re-ID.
- 184
- 183
- Hybrid CNN and Dictionary-Based Models for Scene Recognition and Domain Adaptation168
By integrating the complementary information of MLR, CFV, and the CNN features of the fully connected layer, the state-of-the-art performance can be achieved on scene recognition and DA problems.
- MSDN: Mutually Semantic Distillation Network for Zero-Shot Learning163
A Mutually Semantic Distillation Network (MSDN) is proposed, which progressively distills the intrinsic semantic representations between visual and attribute features for ZSL, leading to new state-of-the-art performances on three popular challenging benchmarks.
- Dynamic Prototype Convolution Network for Few-Shot Semantic Segmentation119
This work proposes a dynamic prototype convolution network (DPCN) to fully capture the aforementioned intrinsic details for accurate FSS, and shows that DPCN yields superior performances under both 1-shot and 5-shot settings.
- 117
- 117
- 111
- 107
- TransZero++: Cross Attribute-Guided Transformer for Zero-Shot Learning99
A cross attribute-guided Transformer network to refine visual features and learn accurate attribute localization for key semantic knowledge representations in ZSL, which achieves the new state-of-the-art results on three golden ZSL benchmarks and on the large-scale ImageNet dataset.
- 84
- 73
- Saliency Guided Inter- and Intra-Class Relation Constraints for Weakly Supervised Semantic Segmentation71
A saliency guided class-agnostic distance module to pull the intra-category features closer by aligning features to their class prototypes and an object guided label refinement module to take a full use of both the segmentation prediction and the initial labels for obtaining superior pseudo-labels.
- LG-CNN: From local parts to global discrimination for fine-grained recognition70
This paper proposes a jointly trained Convolutional Neural Network (CNN) architecture to solve the fine-grained recognition problem without using part and bounding box information and applies this architecture to generic object recognition datasets with superior performance over the directly fine-tuned CNN architecture.
- MVP-Shot: Multi-Velocity Progressive-Alignment Framework for Few-Shot Action Recognition69
A novel Multi-Velocity Progressive-alignment (MVP-Shot) framework to progressively learn and align semantic-related action features at multi-velocity levels and outperforms current state-of-the-art methods on multiple standard few-shot benchmarks.
- Web-Supervised Network with Softly Update-Drop Training for Fine-Grained Visual Classification66
A novel approach to remove irrelevant samples from the real-world web images during training, and only utilize useful images for updating the networks is proposed, which can alleviate the harmful effects caused by irrelevant noisy web images to achieve better performance.
- 63
- 62
- Semantically Meaningful Class Prototype Learning for One-Shot Image Segmentation60
This paper proposes to leverage the multi-class label information during the episodic training to encourage the network to generate more semantically meaningful features for each category and proposes a fused prototype guidance branch for the segmentation of the query image.
- Video super-resolution based on spatial-temporal recurrent residual networks59
The proposed Spatial-Temporal Recurrent Residual Network (STR-ResNet) is able to efficiently reconstruct videos with diversified contents and complex motions, which outperforms the existing video SR approaches and offers new state-of-the-art performances on benchmark datasets.
- 56
- 52
- 46
- 42
- 41
- 40
- Three 3D graphical representations of DNA primary sequences based on the classifications of DNA bases and their applications40
Three 3D graphical representations of DNA primary sequences, which are called RY-curve, MK-curve and SW-curve, based on three classifications of the DNA bases are introduced and similarity among DNA sequences from different species is examined.
- 39
- 34
- Semantic Compression Embedding for Generative Zero-Shot Learning32
A novel Semantic Compression Embedding Guided Generation (SC-EGG) model, which cascades a semantic compression embedding network (SCEN) and an embedding guided generative network (EGGN) and learns a mapping from the class-level semantic space to the dense-semantic visual space, thus improving the discriminability of the synthesized dense- Semantic unseen visual features.
- Discriminative margin-sensitive autoencoder for collective multi-view disease analysis29
A semantic-sensitive autoencoder, where an encoder projects multi-view visual features into the common semantic-aware latent space, and a decoder is exerted as an additional constraint to reconstruct the respective visual features, which can achieve very encouraging performance on protein fold recognition.
- Age progression: Current technologies and applications24
This paper summarizes some recent and impressive age progression methods and gives some applications related to age progression, some popular face aging databases are introduced, and the future directions of the age progression are discussed.
- Foundation Model for Skeleton-Based Human Action Understanding23
A Unified Skeleton-based Dense Representation Learning (USDRL) framework is presented, which serves as a foundational model for skeleton-based human action understanding and significantly outperforms the current state-of-the-art methods.
- Cross Domain Mean Approximation for Unsupervised Domain Adaptation23
A novel discrepancy metric, referred to as Cross Domain Mean Approximation (CDMA) discrepancy, is proposed to evaluate the distribution differences between source and target domains, which calculate the sum of the squares of the distances from the source andtarget domains to the mean of the other domain.
- 21
- SADIH: Semantic-Aware DIscrete Hashing21
A unified Semantic-Aware DIscrete Hashing framework, which aims to directly embed the transformed semantic information into the asymmetric similarity approximation and discriminative hashing function learning and an efficient alternating optimization algorithm is proposed to solve the resulting discrete optimization problem.
- 20
- 20
- Efficient Feature Coding Based on Auto-encoder Network for Image Classification20
This paper proposes to use the Auto-Encoder (AE) network as a local descriptor coding block, and further embed AE network in the BoW framework for the purpose of image classification, and adds an efficient and effective regularization term into the learning process of AE network to promote sparsity of the hidden layer for each input descriptor.
- When SAM2 meets video camouflaged object segmentation: a comprehensive evaluation and adaptation18
This study investigates the application and performance of the Segment Anything Model 2 (SAM2) in the challenging task of video camouflaged object segmentation (VCOS) and demonstrates that SAM2 has the excellent zero-shot ability to detect camouflaged objects in videos.
- Label-Efficient Few-Shot Semantic Segmentation with Unsupervised Meta-Training17
The goal of this paper is to alleviate the training cost for few-shot semantic segmentation models by taking a pioneering step towards label-efficient training of FSS models from fully unlabeled training data, or additionally a few labeled samples to enhance the performance.
- AFANet: Adaptive Frequency-Aware Network for Weakly-Supervised Few-Shot Semantic Segmentation16
This paper utilizes the more challenging image-level annotations and proposes an adaptive frequency-aware network (AFANet) for weakly-supervised few-shot semantic segmentation (WFSS) and proposes a cross-granularity frequency-aware module (CFM) that decouples RGB images into high-frequency and low-frequency distributions and further optimizes semantic structural information by realigning them.
- 16
- 16
- FedMLLM: Federated Fine-tuning MLLM on Multimodal Heterogeneity Data16
A benchmark to evaluate the performance of federated fine-tuning of MLLMs across various multimodal heterogeneous scenarios is introduced and extensive experimental results show that the proposed FL paradigm improves the performance of MLLMs by broadening the range of training data and mitigating multimodal heterogeneity.
- Visual–Semantic Graph Matching Net for Zero-Shot Learning16
A visual-semantic graph matching net (VSGMN), which leverages semantic relationships among classes to aid in visual-semantic embedding and achieves superior performance in both conventional and generalized ZSL scenarios.
- 15
- 15
- Approximately optimizing NDCG using pair-wise loss15
The experimental results from applying the proposed approaches and ten other state-of-the-art methods to five public datasets show the superiority of the proposed methods, especially RankSVMndcg, which are less influenced by the initial weight distribution.
- 14
- Graphical Representation and Similarity Analysis of DNA Sequences Based on Trigonometric Functions14
A 8-component characteristic vector is constructed to compare similarity among DNA sequences from different species based on a normalized geometrical centers of the proposed curves.
- Accelerating Infinite Ensemble of Clustering by Pivot Features14
This work proposes an acceleration version of IEC (AIEC) by extracting the pivot features and learning the multiple mappings to reconstruct them, where the linear equation system can be solved with the time complexity O(dr2) (r ≪ d).
- UnitedVLN: Generalizable Gaussian Splatting for Continuous Vision-Language Navigation13
A novel, generalizable 3DGS-based pre-training paradigm, called UnitedVLN, which enables agents to better explore future environments by unitedly rendering high-fidelity 360 visual images and semantic features, and outperforms state-of-the-art methods on existing VLN-CE benchmarks.
- Learning Anchor Transformations for 3D Garment Animation13
An anchor-based deformation model, namely AnchorDEF, to predict 3D garment animation from a body motion sequence that deforms a garment mesh template by a mixture of rigid transformations with extra nonlinear displacements and an adaptive anchor updating is proposed to optimize the anchor position.
- MUP: Multi-granularity Unified Perception for Panoramic Activity Recognition13
A novel Multi-granularity Unified Perception (MUP) framework that perceives different granularity behaviors universally to explore the co-occurrence motion pattern via the same parameters in an end-to-end fashion is proposed.
- Discriminant Zero-Shot Learning with Center Loss13
This work proposes a new ZSL method with a center loss to make the instances from the same class more compact by extracting their discriminative parts and introduces a varying learning rate to accelerate the model selection process.
- Image Thresholding Segmentation on Quantum State Space13
The method based on global quantum entropy maximization (GQEM) is proposed, which has an equivalent object function to Otsu’s, but gives a more explicit physical interpretation of image thresholding in the language of quantum mechanics.
- LAD-Reasoner: Tiny Multimodal Models are Good Reasoners for Logical Anomaly Detection12
A new task, Reasoning Logical Anomaly Detection (RLAD), which extends traditional anomaly detection by incorporating logical reasoning, and proposes a new framework, LAD-Reasoner, a customized tiny multimodal language model built on Qwen2.5-VL 3B, which excels in producing concise and interpretable rationales.
- 12
- Rethinking attribute localization for zero-shot learning12
A novel attribute localization refinement (ALR) module is proposed designed to enhance the model’s ability to accurately localize all attributes by grouping them and introducing weighted attribute regression to standardize the mapping values of semantic attributes.
- 12
- 11
- 11
- Computational Face Reader10
This paper proposes a novel deep convolutional network in which a facial region pooling layer FRP layer is embedded, called FRP-net and uses the searched facial region windows locates these facial attributes instead of the commonly-used sliding windows.
- USDRL: Unified Skeleton-Based Dense Representation Learning with Multi-Grained Feature Decorrelation9
- Retargeted Multi-View Feature Learning With Separate and Shared Subspace Uncovering9
A multi-view feature learning framework, which can simultaneously learn separate subspace for each view and shared sub space for all the views, respectively, so that each view can preserve the particular information within this view and the shared subspace can capture feature correlation among multiple views.
- 9
- Foreground/Background-Masked Interaction Learning for Spatio-temporal Action Detection8
A novel Foreground/Background-masked Interaction Learning framework to learn the multi-actor features by attentively interacting with the hands-down foreground and background frames and a new Actor-guided Feature Aggregation scheme that integrates foreground- and background-interacted actor features with the learnable actor-based weights are presented.
- 8
- From Class-Specific to Class-Mixture: Cascaded Feature Representations via Restricted Boltzmann Machine Learning8
Two kinds of feature extracting frameworks that can extract cascaded class-specific and class-mixture features, respectively, by taking the restricted Boltzmann machine (RBM) as the basic building blocks are proposed and called CS-RBM and CM-R BM feature extractor are proposed.
- 7
- Neutrosophic Set Transformation Matrix Factorization Based Active Contours for Color Texture Segmentation7
A novel neutrosophic set transformation matrix factorization-based active contour (NSTMF-AC) approach is proposed for color texture segmentation and is compared with the state-of-the-art segmentation algorithms on a variety of natural images.
- Scene classification-oriented saliency detection via the modularized prescription7
Experimental results demonstrate the effectiveness and superiority of the proposed method of scene classification-oriented saliency detection via the modularized prescription in both higher precision and better smoothness, especially for detecting in structure complex scenes.
- 6
- 6
- AdaFPP: Adapt-Focused Bi-Propagating Prototype Learning for Panoramic Activity Recognition6
A novel Adapt-Focused bi-Propagating Prototype learning (AdaFPP) framework to jointly recognize individual, group, and global activities in panoramic activity scenes by learning an adapt-focused detector and multi-granularity prototypes as the pretext tasks in an end-to-end way is proposed.
- 6
- AnomalyControl: Highly-Aligned Anomalous Image Generation with Controlled Diffusion Model5
A controlled anomaly generation model named AnomalyControl is proposed to generate realistic anomalous images aligned highly with both text prompts and anomaly masks that outperform the state-of-the-art anomaly generation methods in terms of authenticity and consistency, and can significantly improve the performance of downstream anomaly detection tasks, especially anomaly localization.
- 5
- 4
- 4
- 3
- PAMD: plausibility-aware motion diffusion model for long dance generation3
Extensive experiments show that PAMD significantly improves musical alignment and enhances the physical plausibility of generated motions, and is incorporated into prior motion guidance (PMG), which uses standing poses as auxiliary conditions alongside music features.
- Hierarchical Vision-Language Interaction for Facial Action Unit Detection3
This work proposes a Hierarchical Vision-language Interaction for AU Understanding (HiVA) method, which leverages textual AU descriptions as semantic priors to guide and enhance AU detection, and enables HiVA to leverage multi-grained vision-based AU features in conjunction with refined language-based AU details, culminating in robust and semantically enriched AU detection capabilities.
- A Conditional Probability Framework for Compositional Zero-Shot Learning3
To enhance object feature learning, this paper incorporates textual descriptors to highlight semantically relevant image regions and jointly optimizing object likelihood and conditional attribute likelihood, which effectively captures compositional dependencies and generalizes well to unseen compositions.
- 3
- 3
- Attack-Augmentation Mixing-Contrastive Skeletal Representation Learning3
In A 2 MC, Attack-Augmentation (Att-Aug) is designed to collaboratively perform targeted and untargeted perturbations of skeletons via attack and augmentation respectively, for generating high-quality hard positive features and hard negative features.
- 2
- Reliable and Diverse Hierarchical Adapter for Zero-shot Video Classification2
This work proposes a reliable and diverse Hierarchical Adapter for zero-shot video classification, which consists of Frame-level Cache Refiner and Video-level Cache Updater, and updates the cache during inference based on the principle of diversity.
- Holistic Prototype Attention Network for Few-Shot VOS2
A holistic prototype attention network (HPAN) is proposed for advancing FSVOS, transferring informative knowledge from seen to unseen classes, and exploiting the holistic information from support images and video frames to achieve support-query semantic consistency and inner-frame temporal consistency.
- Nonconvex matrix completion with Nesterov’s acceleration2
This work adopts the randomized SVD decomposition and Nesterov’s momentum to accelerate the optimization of nonconvex matrix completion and shows that the proposed algorithm achieves tremendous speed increases whilst maintaining comparable performance with the alternating least squares (ALS) method for nonconvergent matrix completions.
- Hybrid of extended locality-constrained linear coding and manifold ranking for salient object detection2
A hybrid method is proposed to ameliorate the fast locating of the foreground while preserving the precision and the noise near the foreground boundary in saliency maps and to generate more accurate saliency.
- Prototype-Guided Curriculum Learning for Zero-Shot Learning1
This work proposes a prototype-guided curriculum learning framework (dubbed as CLZSL), which mitigates instance-level mismatches through a Prototype-Guided Curriculum Learning (PCL) module and addresses class-level imprecision via a Prototype Update (PUP) module.
- 3D-aware Select, Expand, and Squeeze Token for Aerial Action Recognition1
A novel framework, dubbed 3D-Tok, to Select, Expand, and Squeeze original visual tokens for obtaining compact yet diverse semantic-enhanced tokens, which shows significant improvements on the UAV-Human dataset, RoCoG-v2 dataset, and Drone-Action dataset.
- 1
- 1
- Kernelized Similarity Learning and Embedding for Dynamic Texture Synthesis1
This article first put forward two hypotheses that are essential for the DT model to generate new frames using similarity correlations, and integrates kernel learning and the extreme learning machine into a unified synthesis model to learn kernel similarity embeddings for representing DTs.
- 1
- 1
- Fusion linear representation-based classification1
A novel fusion linear representation-based classification method based on the problem of insufficient sample can be solved well by means of mirror image and inverse image.
- –
- –
- –
- –
- –
- –
- –
- –
- –
- Author response for "Multi-Granularity Aggregation Network for Remote Sensing Few-Shot Segmentation"–
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.