Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks89 from public data
- Image-to-image Translation via Hierarchical Style Disentanglement165
Hierarchical Style Disentanglement is proposed, which organizes the labels into a hierarchical tree structure, in which independent tags, exclusive attributes, and disentangled styles are allocated from top to bottom.
- FastVGGT: Training-Free Acceleration of Visual Geometry Transformer108
This work proposes FastVGGT, which, for the first time, leverages token merging in the 3D domain through a training-free mechanism for accelerating VGGT, and devise a unique token partitioning strategy tailored to 3D architectures and tasks.
- You Only Segment Once: Towards Real-Time Panoptic Segmentation86
YOSO is the first real-time panoptic segmentation framework that delivers competitive performance compared to state-of-the-art models and re-parameterizes interpolation-first modules in a convolution-first way, which significantly speeds up the pipeline without any additional costs.
- 84
- 80
- 77
- GOI: Find 3D Gaussians of Interest with an Optimizable Open-vocabulary Semantic-space Hyperplane74
GOI is introduced, a framework that integrates semantic features from 2D vision-language foundation models into 3D Gaussian Splatting (3DGS) and identifies 3D Gaussians of Interest using an Optimizable Semantic-space Hyperplane.
- Cantor: Inspiring Multimodal Chain-of-Thought of MLLM74
It is argued that converging visual context acquisition and logical reasoning is pivotal for tackling visual reasoning tasks and an innovative multimodal CoT framework, termed Cantor, characterized by a perception-decision architecture is proposed, demonstrating the efficacy of the proposed framework.
- Applications and challenges of biomarker-based predictive models in proactive health management55
This systematic approach enhances early disease screening accuracy while supporting risk stratification and precision diagnosis, particularly for chronic conditions and oncology applications, by effectively connecting biomarker discovery with practical clinical utilization.
- Scoot: A Perceptual Metric for Facial Sketches52
The results suggest that “spatial structure” and “co-occurrence texture” are two generally applicable perceptual features in face sketch synthesis, and the first largest scale human-perception-based sketch database that can evaluate how well a metric consistent with human perception is evaluated.
- Robust Face Sketch Style Synthesis52
This work proposes a robust face sketch style synthesis method, which can convert photos to arbitrarily stylistic sketches based on only one corresponding template sketch, and introduces a multi-feature-based optimization model.
- 47
- DistilPose: Tokenized Pose Regression with Heatmap Distillation44
A novel human pose estimation framework termed DistilPose, which bridges the gaps between heatmap-based and regression-based methods and maximizes the transfer of knowledge from the teacher model to the student model through Token-distilling Encoder (TDE) and Simulated Heatmaps.
- Pseudo-label Alignment for Semi-supervised Instance Segmentation40
In PAIS, a dynamic aligning loss (DALoss) that adjusts the weights of semi-supervised loss terms with varying class and mask score pairs is devised that adjusts the weights of semi-supervised loss terms with varying class and mask score pairs.
- InterFormer Real-time Interactive Image Segmentation40
The effectiveness of Inter-Former is demonstrated, which outperforms previous interactive segmentation models in terms of computational efficiency and segmentation quality, achieve real-time high-quality interactive segmentation on CPU-only devices.
- Dense Auto-Encoder Hashing for Robust Cross-Modality Retrieval40
This paper proposes a novel cross-modality hashing scheme, termed Dense Auto-encoder Hashing (DAH), which can explicitly impute the missed modality and produce robust binary codes by leveraging the relatedness among different modalities.
- Robust Face Sketch Synthesis via Generative Adversarial Fusion of Priors and Parametric Sigmoid39
A novel generative adversarial network termed pGAN is proposed, which can generate face sketches efficiently using training data under fixed conditions and handle the aforementioned uncontrolled conditions.
- Drag Your Gaussian: Effective Drag-Based Editing with Score Distillation for 3D Gaussian Splatting33
DYG, a 3D drag-based editing framework for 3DGS that integrates the implicit triplane representation to establish the geometric scaffold of editing results, effectively overcoming suboptimal editing outcomes caused by the sparsity of 3DGS in the desired editing regions.
- FlashWorld: High-quality 3D Scene Generation within Seconds33
This work proposes FlashWorld, a generative model that produces 3D scenes from a single image or text prompt in seconds, 10~100$\times$ faster than previous works while possessing superior rendering quality and proposes a cross-mode post-training distillation that enhances visual quality while maintaining 3D consistency.
- 33
- Multi-modal Multi-layer Fusion Network with Average Binary Center Loss for Face Anti-spoofing32
A novel Multi-modal Multi-layer Fusion Convolutional Neural Network (mmfCNN), which targets at finding a discriminative model for recognizing the subtle differences between live and spoof faces, and utilizes a multi-layer fusion model to further aggregate the features from different layers.
- 31
- Adaptive Selection based Referring Image Segmentation25
An innovative framework for RIS that seeks to overcome challenges with adaptive alignment of vision and language features, termed the Adaptive Selection with Dual Alignment (ASDA), which exceeds state-of-the-art methods on RefCOCO, RefCOCO+ and G-Ref benchmark.
- 23
- 23
- Attack Can Benefit: An Adversarial Approach to Recognizing Facial Expressions under Noisy Annotations21
A novel and flexible method to spot noisy labels by leveraging adversarial attack, termed as Geometry Aware Adversarial Vulnerability Estimation (GAAVE), which has no reliance on additional information and is thus easy to generalize to the large-scale real-world FER datasets.
- Unified framework for face sketch synthesis21
This work proposes a simple yet effective unified approach considering position constraint, global search and high frequency compensation to infer pseudosketches from input test photos to achieve superior performance in comparison to state-of-the-art face sketch synthesis methods.
- Joint segmentation and detection of COVID-19 via a sequential region generation network20
A simple yet efficient and general-purpose network, called Sequential Region Generation Network (SRGNet), to jointly detect and segment the lesion areas of COVID-19 and generate high-quality lesion-areas suggestions on the predicted segmentation maps, reducing the diagnosis cost.
- 19
- 17
- 17
- Fast Face Sketch Synthesis via KD-Tree Search17
A simple yet effective fast face sketch synthesis method based on K dimensional-Tree (KD-Tree), which employs the idea of divide-and-conquer to learn the complex nonlinear mapping between facial photos and sketches.
- EOV-Seg: Efficient Open-Vocabulary Panoptic Segmentation16
EOV-Seg, a novel single-stage, shared, efficient, and spatialaware framework designed for open-vocabulary panoptic segmentation, is proposed, which runs faster and achieves competitive performance compared with state-of-the-art methods.
- 15
- Towards Visual Feature Translation15
This paper proposes a Hybrid Auto-Encoder (HAE) to translate visual features, which learns a mapping by minimizing the translation and reconstruction errors, and an Undirected Affinity Measurement (UAM) is further designed to quantify the affinity among different types of visual features.
- DMAD: Dual Memory Bank for Real-World Anomaly Detection14
The results show that DMAD surpasses current state-of-the-art methods, highlighting DMAD's capability in handling the complexities of real-world anomaly detection scenarios.
- UCOD-DPL: Unsupervised Camouflaged Object Detection via Dynamic Pseudo-label Learning13
A UCOD method with a teacher-student framework via Dynamic Pseudo-label Learning called UCOD-DPL, which contains an Adaptive Pseudo-label Module (APM), a Dual-Branch Adversarial (DBA) decoder, and a Look-Twice mechanism, which demonstrates outstanding performance.
- Training-Free Hierarchical Scene Understanding for Gaussian Splatting with Superpoint Graphs13
This work introduces a training-free framework that constructs a superpoint graph directly from Gaussian primitives that ensures strong 3D semantic coherence and naturally supports hierarchical understanding, enabling both coarse- and fine-grained open-vocabulary perception within a unified semantic field.
- Evolving High-Quality Rendering and Reconstruction in a Unified Framework with Contribution-Adaptive Regularization12
The essence of the framework is learning adaptive contribution for Gaussian primitives by squeezing the knowledge from geometry regularization into a compact MLP, which improves the mutual learning of the two tasks, and its unified structure doesn’t require separate models as in dual-model based approaches, guaranteeing efficiency.
- Dual3D: Efficient and Consistent Text-to-3D Generation with Dual-mode Multi-view Latent Diffusion12
The dual-mode toggling inference strategy is proposed to use only $1/10$ denoising steps with 3D mode, successfully generating a 3D asset in just $10$ seconds without sacrificing quality.
- 12
- CAKE: Category Aware Knowledge Extraction for Open-Vocabulary Object Detection11
A Category Aware Knowledge Extraction framework (CAKE), which consists of a Category-Specific Knowledge Distillation branch (CSKD) and a Category Generalization Region Proposal Network (CG-RPN) that leverages the guidance of feature set to adjust the confidence scores of region proposals, thereby mining proposals that potentially contain novel categories of objects.
- 11
- Geometric-aware Pretraining for Vision-centric 3D Object Detection11
This work proposes a novel geometric-aware pretraining framework called GAPretrain, which incorporates spatial and structural cues to camera networks by employing the geometric-rich modality as guidance during the pretraining phase and serves as a plug-and-play solution that can be flexibly applied to multiple state-of-the-art detectors.
- 11
- 9
- FocSAM: Delving Deeply into Focused Objects in Segmenting Anything8
FocSAM augments SAM's interactive segmentation performance to match the existing state-of-the-art method in segmentation quality, requiring only about 5.6% of this method's inference time on CPUs.
- 8
- Face sketch aging via aging oriented principal component analysis8
A simple but effective method for face sketch aging is proposed that escapes the process of collecting sketch aging sequences elegantly and results validate that the proposed method is effective and promising.
- EBFT: Effective and Block-Wise Fine-Tuning for Sparse LLMs7
This work proposes an efficient and fast framework for fine-tuning sparse LLMs based on minimizing reconstruction error that achieves superior results on various benchmarks, and can be executed on a single 16GB GPU.
- 6
- LiDAR-guided Geometric Pretraining for Vision-Centric 3D Object Detection6
This work proposes a novel geometric-aware pretraining framework called GAPretrain, which incorporates spatial and structural cues to camera networks by employing the geometric-rich modality as guidance during the pretraining phase and serves as a plug-and-play solution that can be flexibly applied to multiple state-of-the-art detectors.
- BUFF: Bayesian Uncertainty Guided Diffusion Probabilistic Model for Single Image Super-Resolution6
The Bayesian Uncertainty Guided Diffusion Probabilistic Model (BUFF) is introduced, incorporating a Bayesian network to generate high-resolution uncertainty masks that guide the diffusion process and significantly mitigates artifacts and blurring in areas characterized by complex textures and fine details.
- 6
- CAM R-CNN: End-to-End Object Detection with Class Activation Maps6
CAM R-CNN is proposed, in which the category-aware attention maps provided by CAMs are integrated into the process of object detection, and its superiority compared to several strong baselines on the challenging COCO dataset is demonstrated.
- DeOcc-1-to-3: 3D De-Occlusion from a Single Image via Self-Supervised Multi-View Diffusion5
DeOcc-1-to-3, an end-to-end framework for occlusion-aware multi-view generation, and a self-supervised training pipeline that leverages occluded–unoccluded image pairs and pseudo-ground-truth views to guide structure-aware completion and view consistency.
- XSpecMesh: Quality-Preserving Auto-Regressive Mesh Generation Acceleration via Multi-Head Speculative Decoding5
This work introduces XSpecMesh, a quality-preserving acceleration method for auto-regressive mesh generation models, which employs a lightweight, multi-head speculative decoding scheme to predict multiple tokens in parallel within a single forward pass, thereby accelerating inference.
- Multi-Modal Prompt Learning on Blind Image Quality Assessment5
This paper introduces an innovative multi-modal prompt-based methodology for IQA that employs carefully crafted prompts that synergistically mine incremental semantic information from both visual and linguistic data.
- Face Sketch Recognition via Data-Driven Synthesis5
This chapter would give a comprehensive analysis and comparison to advances on face sketch–photo synthesis methods and the probe sketch drawn by the artist is matched with the transformed sketches subsequently.
- 4
- SynergyAmodal: Deocclude Anything with Text Control3
SynergyAmodal is proposed, a novel framework for co-synthesizing in-the-wild amodal datasets with comprehensive shape and appearance annotations, which integrates these elements through a tripartite data-human-model collaboration.
- NeRF-DetS: Enhanced Adaptive Spatial-wise Sampling and View-wise Fusion Strategies for NeRF-based Indoor Multi-view 3D Object Detection3
The proposed NeRF-DetS includes two modules: Progressive Adaptive Sampling Strategy (PASS) and Depth-Guided Simplified Multi-Head Attention Fusion (DS-MHA), which can automatically sample features of each layer within a dense 3D detector, using offsets predicted by the previous layer.
- SCOUT: Semi-supervised Camouflaged Object Detection by Utilizing Text and Adaptive Data Selection3
This work introduces a Semi-supervised Camouflaged Object Detection by Utilizing Text and Adaptive Data Selection (SCOUT), which surpasses previous semi-supervised methods in the COD field and achieves state-of-the-art performance.
- Generate Aligned Anomaly: Region-Guided Few-Shot Anomaly Image-Mask Pair Synthesis for Industrial Inspection3
Generate Aligned Anomaly (GAA), a region-guided, few-shot anomaly image-mask pair generation framework that leverages the strong priors of a pretrained latent diffusion model to generate realistic, diverse, and semantically aligned anomalies using only a small number of samples.
- SEA-Vision: A Multilingual Benchmark for Comprehensive Document and Scene Text Understanding in Southeast Asia2
SEA-Vision, a benchmark that jointly evaluates Document Parsing and Text-Centric Visual Question Answering (TEC-VQA) across 11 Southeast Asian languages, is introduced and a hybrid pipeline for Document Parsing and TEC-VQA is designed, which combines automated filtering and scoring with MLLM-assisted labeling and lightweight native-speaker verification.
- What You Perceive Is What You Conceive: A Cognition-Inspired Framework for Open Vocabulary Image Segmentation2
A novel Cognition-Inspired Framework for open vocabulary image segmentation that emulates the human visual recognition process: first forming a conceptual understanding of an object, then perceiving its spatial extent, offering enhanced flexibility in recognizing unseen categories.
- 2
- Breaking the Bias: Recalibrating the Attention of Industrial Anomaly Detection2
Recalibrating Attention of Industrial Anomaly Detection (RAAD) is proposed, a framework that systematically decomposes and recalibrates attention maps and optimizes both computational efficiency and the model' s sensitivity to anomalies.
- 2
- AAM Based Face Sketch Synthesis2
This paper introduces AAM to “explain” face photos by generating synthetic images that are as similar as possible and provides a compact set of parameters that are useful for face sketch synthesis.
- A Heterogeneous Image Transformation Based Synthesis Framework for Face Sketch Aging2
A novel heterogeneous image transformation (HIT) based synthesis framework for face sketch aging is proposed, where face sketches to be aged are first transferred to pseudo-photos by existing HIT methods, then existing face aging methods are employed to obtain corresponding aged pseudo- photos.
- S²Teacher: Step-by-step Teacher for Sparsely Annotated Oriented Object Detection1
The S2Teacher is proposed, a novel angle-consistency guided method that progressively mines pseudo-labels for unlabeled objects from easy to hard, enhancing foreground representations and balancing accuracy and labeling cost.
- Robust Pseudo-Labeling via Decoupled Class-Aware Filtering and Dynamic Category Correction1
A novel Pseudo-Label Decoupling and Correction (PL-DC) framework is proposed, which explicitly decouples and enhances the pseudo-label selection process for SSIS, and designs a Dynamic Instance Category Correction module that reassigns ambiguous class pseudo-label by leveraging semantic prototypes and consistency alignment.
- 1
- 1
- Probabilistic Automated Model Compression via Representation Mutual Information Optimization1
This work introduces a probabilistic framework for automated model compression (Prob-AMC) that optimizes pruning, quantization, and knowledge distillation simultaneously using information theory, and achieves a superior compression ratio on CIFAR-10 and ImageNet.
- Edge reconstruction and feature enhancement‐driven architecture for blind super‐resolution in medical imaging systems1
This study introduces a novel degradation comparative learning framework meticulously designed for the nuanced degradation characteristics of medical images within the Internet of Medical Things (IoMT), underpinned by edge reconstruction and an innovative image feature supplement module.
- 1
- 1
- Backdoor Attack against Face Sketch Synthesis1
This article proposes a simple yet effective poison-only backdoor attack suitable for generation tasks in facial sketch synthesis and demonstrates that when the backdoor is integrated into the target model via the attack, it can mislead the model to synthesize unacceptable sketches of any photos stamped with the trigger patterns.
- –
- –
- –
- –
- –
- –
- –
- –
- Direction based Graphical Model for Face Sketch Synthesis–
A direction based graphical model (DGM) is proposed to improve face sketch synthesis by combining image directions with image intensities and Markov network to infer the synthesized sketch.
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.