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
- Dacheng TaoSuggested from co-authorship
Is this you? Claim this profile to confirm or dismiss it.
Works89 from public data
- Robust Face Recognition via Multimodal Deep Face Representation460
A comprehensive deep learning framework to jointly learn face representation using multimodal information is proposed and a small ensemble system achieves higher than 99.0% recognition rate on LFW using publicly available training set.
- Trunk-Branch Ensemble Convolutional Neural Networks for Video-Based Face Recognition447
A Trunk-Branch Ensemble CNN model (TBE-CNN), which extracts complementary information from holistic face images and patches cropped around facial components, achieves state-of-the-art performance on three popular video face databases: PaSC, COX Face, and YouTube Faces.
- Multi-Directional Multi-Level Dual-Cross Patterns for Robust Face Recognition387
Experimental results indicate that DCP outperforms the state-of-the-art local descriptors for both face identification and face verification tasks and the best performance is achieved on the challenging LFW and FRGC 2.0 databases by deploying MDML-DCPs in a simple recognition scheme.
- GPS-Net: Graph Property Sensing Network for Scene Graph Generation348
A novel message passing module that augments the node feature with node-specific contextual information and encodes the edge direction information via a tri-linear model is proposed that achieves state-of-the-art performance on three popular databases: VG, OI, and VRD by significant gains under various settings and metrics.
- Two-Stage Cascaded U-Net: 1st Place Solution to BraTS Challenge 2019 Segmentation Task340
A novel two-stage cascaded U-Net to segment the substructures of brain tumors from coarse to fine is devised and won the 1st place in the BraTS 2019 challenge segmentation task.
- A Comprehensive Survey on Pose-Invariant Face Recognition330
The inherent difficulties in PIFR are discussed and a comprehensive review of established techniques are presented, that is, pose-robust feature extraction approaches, multiview subspace learning approaches, face synthesis approaches, and hybrid approaches.
- 269
- Dual-force convolutional neural networks for accurate brain tumor segmentation197
Results on the two databases indicate that the proposed approaches consistently promote the segmentation performance of the two popular deep models, BRATS 2017 and BRATS 2015.
- Learning Granularity-Unified Representations for Text-to-Image Person Re-identification196
An end-to-end framework based on transformers to learn granularity-unified representations for both modalities, denoted as LGUR, which consistently outperforms state-of-the-arts by large margins on both CUHK-PEDES and ICFG-P EDES datasets.
- Pose-invariant face recognition with homography-based normalization128
An highly efficient PIFR algorithm that effectively handles the main challenges caused by pose variation and an effective approach for occlusion detection, which enables face recognition with visible patches only is proposed.
- Glance and Gaze: Inferring Action-aware Points for One-Stage Human-Object Interaction Detection123
This paper proposes a novel one-stage method, namely Glance and Gaze Network (GGNet), which adaptively models a set of action-aware points (ActPoints) via glance and gaze steps, and outperforms state-of-the-art methods by significant margins on both V-COCO and HICO-DET benchmarks.
- Learning Contextual and Attentive Information for Brain Tumor Segmentation118
This work designs multiple deep architectures of varied structures to learning contextual and attentive information, then ensemble the predictions of these models to obtain more robust segmentation results, and shows that the risk of overfitting in segmentation is reduced.
- CPP-Net: Context-Aware Polygon Proposal Network for Nucleus Segmentation109
A Context-aware Polygon Proposal Network (CPP-Net) for nucleus segmentation that substantially enhances contextual information and thereby improves the robustness of the prediction and is found to achieve state-of-the-art performance on three publicly available databases.
- Correcting the Triplet Selection Bias for Triplet Loss107
A new variant of triplet loss is proposed, which tries to reduce the bias in triplet selection by adaptively correcting the distribution shift on the selected triplets.
- Multi-Task Learning With Coarse Priors for Robust Part-Aware Person Re-Identification104
A robust, compact, and easy-to-use method called the Multi-task Part-aware Network (MPN), which is designed to extract semantically aligned part-level features from pedestrian images and consistently outperforms state-of-the-art approaches by significant margins.
- Quality-Aware Part Models for Occluded Person Re-Identification101
A novel method named Quality-aware Part Models (QPM) for occlusion-robust ReID, which does not rely on any outside tools in either the training or inference stages, and handles occlusions caused by both objects and other pedestrians; it is highly computationally efficient.
- Harnessing the Power of MLLMs for Transferable Text-to-Image Person ReID91
A novel method is introduced that automatically identifies words in a description that do not correspond with the image, based on the similarity between one text and all patch token embeddings in the image, which alleviates the impact of noisy textual descriptions.
- MobileFaceSwap: A Lightweight Framework for Video Face Swapping87
A lightweight Identity-aware Dynamic Network (IDN) is proposed for subject-agnostic face swapping by dynamically adjusting the model parameters according to the identity information by introducing two dynamic neural network techniques, including the weights prediction and weights modulation.
- One-Pass Multi-task Convolutional Neural Networks for Efficient Brain Tumor Segmentation86
This paper decompose brain tumor segmentation into three different but related tasks and proposes a multi-task deep model that trains them together to exploit their underlying correlation and introduces a simple yet effective post-processing method that can further improve the segmentation performance significantly.
- 83
- Unified Pre-training with Pseudo Texts for Text-To-Image Person Re-identification80
A new unified pre-training pipeline (UniPT) designed specifically for the T2I-ReID task is presented, in which pseudo-textual descriptions of images are automatically generated by the CLIP paradigm using a divide-conquer-combine strategy.
- 79
- Polysemy Deciphering Network for Human-Object Interaction Detection73
A novel Polysemy Deciphering Network (PD-Net), which decodes the visual polysemy of verbs for HOI detection in three ways, and achieves the best performance on both HICO-DET and V-COCO datasets.
- Polysemy Deciphering Network for Robust Human–Object Interaction Detection66
A novel Polysemy Deciphering Network (PD-Net) is proposed that decodes the visual polysemy of verbs for HOI detection in three distinct ways and is demonstrated to outperform state-of-the-art methods by significant margins on the HICO-DET, V-COCO, and HOI-VP databases.
- Texture-Preserving Diffusion Models for High-Fidelity Virtual Try-On64
An Texture-Preserving Diffusion (TPD) model for virtual try-on is proposed, which enhances the fidelity of the results and introduces no additional image encoders, and a novel diffusion-based method is proposed that predicts a precise inpainting mask based on the person and reference garment images, further enhancing the reliability of the try-on results.
- 61
- HL-Net: Heterophily Learning Network for Scene Graph Generation54
A novel Heterophily Learning Network (HL-Net) is proposed to comprehensively explore the homophily and heterophily be-tween objects/relationships in scene graphs and demonstrates the superiority of the proposed HL-Net over existing state-of-the-art approaches.
- Uncertainty-Aware Clustering for Unsupervised Domain Adaptive Object Re-Identification52
A novel hierarchical clustering scheme is proposed to promote clustering quality, and an uncertainty-aware collaborative instance selection method is introduced to select images with reliable labels for model training that effectively reduces the impact of noisy labels.
- Soft-Ranking Label Encoding for Robust Facial Age Estimation52
A novel age encoding method, referred to as ‘Soft-ranking’, which encodes two important properties of facial age, the ordinal property and the correlation between adjacent ages, which provides a richer supervision signal for training deep models.
- Trajectory Consistency Distillation: Improved Latent Consistency Distillation by Semi-Linear Consistency Function with Trajectory Mapping51
Trajectory Consistency Distillation (TCD) is introduced, which encompasses trajectory consistency function and strategic stochastic sampling and not only significantly enhances image quality at low NFEs but also yields more detailed results compared to the teacher model at high NFEs.
- 51
- Towards Variable and Coordinated Holistic Co-Speech Motion Generation50
This paper addresses the problem of generating lifelike holistic co-speech motions for 3D avatars, focusing on two key aspects: variability and coordination, with ProbTalk, a unified probabilistic framework designed to jointly model facial, hand, and body movements in speech.
- Style Interleaved Learning for Generalizable Person Re-Identification50
This work proposes a novel style interleaved learning (IL) framework that not only consistently outperforms state-of-the-art methods on large-scale benchmarks for DG ReID, but also has clear advantages in computational efficiency.
- RU-Net: Regularized Unrolling Network for Scene Graph Generation48
This paper first study the relation between GMP and graph Laplacian denoising (GLD) from the perspective of the unrolling technique, determining that GMP can be formulated as a solver for GLD, and proposes an unrolled message passing module and an fp-based graph regularization to suppress spurious connections between nodes.
- Learning Oracle Attention for High-Fidelity Face Completion47
This paper designs a comprehensive framework for face completion based on the U-Net structure and proposes a dual spatial attention module to efficiently learn the correlations between facial textures at multiple scales, which outperforms state-of-the-art methods by large margins.
- Boundary-Assisted Region Proposal Networks for Nucleus Segmentation45
A Boundary-assisted Region Proposal Network (BRP-Net) that achieves robust instance-level nucleus segmentation through a novel Task-aware Feature Encoding (TAFE) network that efficiently extracts respective high-quality features for semantic segmentation and instance boundary detection tasks.
- 44
- 44
- Can Linguistic Knowledge Improve Multimodal Alignment in Vision-Language Pretraining?32
This study aims to shed light on the influence of comprehensive linguistic knowledge encompassing semantic expression and syntactic structure on multimodal alignment in vision-language pretraining by introducing SNARE, a large-scale multimodal alignment probing benchmark designed specifically for the detection of vital linguistic components.
- Few-Shot Head Swapping in the Wild32
This paper presents the Head Swapper (HeSer), which achieves few-shot head swapping in the wild through two delicately de-signed modules, and demonstrates that the proposed method produces superior head swapping results on a variety of scenes.
- Guiding Human-Object Interactions with Rich Geometry and Relations28
ROG is introduced, a novel diffusion-based framework that models the spatiotemporal relationships inherent in HOIs with rich geometric detail and significantly outperforms state-of-the-art methods in the realism and semantic accuracy of synthesized HOIs.
- HelixSurf: A Robust and Efficient Neural Implicit Surface Learning of Indoor Scenes with Iterative Intertwined Regularization28
A method termed Helix-shaped neural implicit Surface learning or HelixSurf, which uses the intermediate prediction from one strategy as the guidance to regularize the learning of the other one, and conducts such intertwined regularization iteratively during the learning process.
- 28
- 26
- Disentangled Pre-Training for Human-Object Interaction Detection20
The DP-HOI structure can be easily adapted to the HOI detection task, enabling effective model parameter initialization and significantly enhances the performance of existing HOI detection models on a broad range of rare categories.
- Modeling Thousands of Human Annotators for Generalizable Text-to-Image Person Re-identification19
A Human Annotator Modeling (HAM) approach is proposed to enable MLLMs to mimic the description styles of thousands of human annotators and is adopted to automatically annotate a massive-scale database for text-to-image ReID.
- 17
- Robust face recognition17
This thesis addresses key challenges inherent in real-world face recognition systems including pose and illumination variations, occlusion, and image blur, and develops a series of robust face recognition algorithms proposed.
- ChartMaster: Advancing Chart-to-Code Generation with Real-World Charts and Chart Similarity Reinforcement Learning15
The ChartMaster model, a GRPO-based reinforcement learning algorithm guided by a novel chart similarity reward, is developed, achieving SOTA results among 7B-parameter models and rivaling GPT-4o on various chart-to-code benchmarks.
- Context Sensing Attention Network for Video-based Person Re-identification13
This work proposes a novel Context Sensing Attention Network (CSA-Net), which improves both the frame feature extraction and temporal aggregation steps and introduces the Contrastive Feature Aggregation (CFA) module, which predicts frame weights for temporal aggregation.
- Decoupled Prototype Learning for Reliable Test-Time Adaptation12
This study reveals that minimizing the classification error of each sample causes the cross-entropy loss's vulnerability to label noise to be exposed.
- Local-consistent Transformation Learning for Rotation-invariant Point Cloud Analysis12
Equipped with LCRF and RPR, the LocoTrans is capable of learning local-consistent transformation and preserving local geometry, which benefits rotation invariance learning.
- Simultaneous Computation and Memory Efficient Zeroth-Order Optimizer for Fine-Tuning Large Language Models10
A novel layer-wise sparse computation and memory efficient ZO optimizer, named LeZO, which accelerates training without compromising the performance of ZO optimization and incorporates layer-wise parameter sparsity in the process of simultaneous perturbation stochastic approximation and ZO stochastic gradient descent.
- 10
- Classification of single-view object point clouds10
It is shown in this paper that performance of existing point cloud classifiers drops drastically under the considered single-view, partial setting; the phenomenon is consistent with the observation that semantic category of a partial object surface is less ambiguous only when its distribution on the whole surface is clearly specified.
- Attention-guided Progressive Mapping for Profile Face Recognition10
This paper proposes to learn the feature residual between the source pose and its nearby pose in a block-by-block fashion, and thus traversing to the feature space of a smaller pose by adding the learned residual.
- Effortless Active Labeling for Long-Term Test-Time Adaptation9
This paper investigates how to achieve effortless active labeling so that a maximum of one sample is selected for annotation in each batch, and introduces an efficient strategy to identify these samples using feature perturbation.
- Bilateral Collaboration with Large Vision-Language Models for Open Vocabulary Human-Object Interaction Detection8
A novel Bilateral Collaboration framework for open vocabulary HOI detection (BC-HOI) is proposed, which includes an Attention Bias Guidance (ABG) component, which guides the VLM to produce fine-grained instance-level interaction features according to the attention bias provided by the HOI detector.
- 8
- 7
- MMoT: Mixture-of-Modality-Tokens Transformer for Composed Multimodal Conditional Image Synthesis7
A Mixture-of-Modality-Tokens Transformer (MMoT) that adaptively fuses fine-grained multimodal control signals, a multimodal balanced training loss to stabilize the optimization of each modality, and a multimodal sampling guidance to balance the strength of each modality control signal is introduced.
- Cross-modal & Cross-domain Learning for Unsupervised LiDAR Semantic Segmentation6
The experimental results on several datasets show that in the proposed setting, the developed CoMoDaL can achieve segmentation without the supervision of labeled LiDAR data.
- Learning Context with Priors for 3D Interacting Hand-Object Pose Estimation5
A novel and robust approach that learns a broad range of context by imposing priors that significantly outperforms state-of-the-art approaches and can be applied to other hand pose estimation tasks.
- Mining False Positive Examples for Text-Based Person Re-Identification5
This work delicately design a cross-relu loss to amplify the dissimilarity between matched and mismatched word-region pairs, effectively preventing false-positive matching in text-based person re-identification.
- Decoupling Heterogeneous Features for Robust 3D Interacting Hand Poses Estimation4
This work presents a novel framework called Decoupled Feature Learning (DFL) for 3D pose estimation of interacting hands that significantly outperforms state-of-the-art methods on the large-scale InterHand2.6M dataset and exhibits strong generalization ability on in-the-wild images.
- 4
- ViHOI: Human-Object Interaction Synthesis with Visual Priors3
ViHOI is introduced, a novel framework that enables diffusion-based generative models to leverage rich, task-specific priors from 2D images to enhance generation quality and improve generalization to unseen objects and interaction categories.
- Harnessing Text-to-Image Diffusion Models for Point Cloud Self-Supervised Learning3
Comprehensive experiments on downstream point cloud tasks and ablation studies demonstrate that the SD model can enhance point cloud self-supervised learning.
- 3
- Streamlined Open-Vocabulary Human-Object Interaction Detection2
SL-HOI is introduced, a StreamLined open-vocabulary HOI detection framework based solely on the powerful DINOv3 model that achieves state-of-the-art performance on both the SWiG-HOI and HICO-DET benchmarks, demonstrating the effectiveness of the streamlined model architecture.
- From Answers to Rationales: Self-Aligning Multimodal Reasoning with Answer-Oriented Chain-of-Thought2
Experiments indicate that the SMART framework significantly improves various MLLMs, regardless of model architecture, parameter size, or pre-training dataset, and establishes an iterative generation-optimization method that continually enhances the model's reasoning skills.
- 2
- AFPN: Attention-guided Feature Partition Network for Cross-view Geo-localization2
An attention-guided feature partition network (AFPN) is proposed which leverages learnable spatial attention maps to divide the global high-level feature map into the class-aware foreground and theclass-agnostic background feature in an end-to-end learning manner.
- Disentangled Interaction Representation for One-Stage Human-Object Interaction Detection2
This paper proposes Shunted Cross-Attention (SCA) and introduces the Interaction-aware Pose Estimation (IPE) task to learn interaction-relevant human pose features using a disentangled decoder and fuses the appearance feature and pose feature via element-wise addition to form the interaction representation.
- CDPM: Convolutional Deformable Part Models for Person Re-identification.2
Convolutional Deformable Part Models (CDPM) works by decoupling the complex part alignment procedure into two easier steps, and is able to accurately align flexible body parts without the need of any outside information.
- 2
- Coordinating Multiple Conditions for Trajectory-Controlled Human Motion Generation1
Experiments on HumanML3D and KIT datasets demonstrate that CMC achieves state-of-the-art performance in control accuracy and motion quality, demonstrating its effectiveness in coordinating multimodal conditions and representations.
- Beyond Activation Alignment:The Alignment-Diversity Tradeoff in Task-Aware LLM Quantization1
TASA (Task-Aware Sensitivity Analysis), a two-level framework that jointly optimizes calibration-data composition and mixed-precision bit allocation, is proposed and results show that calibration-data composition substantially affects task-sensitive quantization.
- Democratizing High-Fidelity Co-Speech Gesture Video Generation1
A lightweight framework that utilizes 2D full-body skeletons as an efficient auxiliary condition to bridge audio signals with visual outputs and exceeds state-of-theart approaches in visual quality and synchronization while generalizing across speakers and contexts is proposed.
- 1st Place Solution to the EPIC-Kitchens Action Anticipation Challenge 20221
This paper develops Anticipation Time Knowledge Distillation using the soft labels learned by the teacher model as knowledge to guide the student network to learn the information of anticipation time and Verb-Noun Relation Module for building the relationship between verbs and nouns.
- 1
- 1
- –
- –
- –
- –
- Beyond Human Data: Aligning Multimodal Large Language Models by Iterative Self-Evolution–
This work proposes a novel multimodal self-evolution framework that empowers the model to autonomously generate high-quality questions and answers using only unannotated images and designs an answer self-enhancement technique to boost the discriminative power of answers.
- –
- –
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 2026-10-12. 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.