arXivSub Start free trial

CVPR 2025 Papers with Code β€” Page 2

IEEE/CVF Conference on Computer Vision and Pattern Recognition Β· 851 papers

CAV-MAE Sync: Improving Contrastive Audio-Visual Mask Autoencoders via Fine-Grained Alignment

Edson Araujo (Goethe University of Frankfurt), Hilde Kuehne (Tuebingen AI Center University of Tuebingen)

CodeClassificationSegmentationRetrievalTransformerAuto EncoderContrastive LearningMultimodalityAudio

🎯 What it does: Proposes CAV-MAE Sync, which improves audio-visual self-supervised learning through fine-grained audio-visual alignment, separation of contrastive and reconstruction objectives, and the introduction of global and register tokens.

CCIN: Compositional Conflict Identification and Neutralization for Composed Image Retrieval

Likai Tian (Wuhan University), Xuelong Li (Northwestern Polytechnical University)

CodeRetrievalTransformerLarge Language ModelContrastive LearningImage

🎯 What it does: This paper proposes a method to improve image retrieval accuracy by first identifying attribute conflicts between the reference image and modification instructions, and then neutralizing these conflicts through a dual-instruction mechanism.

CDI: Copyrighted Data Identification in Diffusion Models

Jan DubiΕ„ski (Warsaw University of Technology), Adam Dziedzic (CISPA Helmholtz Center for Information Security)

CodeRecognitionData-Centric LearningTransformerDiffusion modelImage

🎯 What it does: This paper proposes a copyright data identification framework called CDI based on dataset inference, which is used to determine whether a Diffusion model has utilized a specific copyrighted dataset.

CGMatch: A Different Perspective of Semi-supervised Learning

Bo Cheng (Jilin University), Lan Du (Monash University)

CodeClassificationOptimizationSupervised Fine-TuningImage

🎯 What it does: This paper proposes a new semi-supervised learning framework called CGMatch, focusing on model training in scenarios with few labeled samples.

Chain of Attack: On the Robustness of Vision-Language Models Against Transfer-Based Adversarial Attacks

Peng Xie (Hong Kong University of Science and Technology), Kani Chen (Hong Kong University of Science and Technology)

CodeAdversarial AttackTransformerLarge Language ModelVision Language ModelContrastive LearningImageTextMultimodality

🎯 What it does: Evaluate the robustness of VLM in black-box transfer adversarial attacks, propose the Chain of Attack (CoA) method to generate adversarial images, and introduce LLM-based ASR automatic evaluation metrics.

Channel Consistency Prior and Self-Reconstruction Strategy Based Unsupervised Image Deraining

Guanglu Dong (Sichuan University), Chao Ren (Beijing Jiaotong University)

CodeImage TranslationRestorationGenerative Adversarial NetworkImage

🎯 What it does: This paper proposes an unsupervised single image de-raining framework called CSUD, which eliminates the dependence on real rain-clean paired data through pseudo-pair generation and self-reconstruction strategies.

Charm: The Missing Piece in ViT Fine-Tuning for Image Aesthetic Assessment

Fatemeh Behrad (KU Leuven University), Johan Wagemans (KU Leuven University)

CodeClassificationComputational EfficiencyTransformerSupervised Fine-TuningImage

🎯 What it does: A new tokenization method called Charm is proposed, which retains the composition, high definition, aspect ratio, and multi-scale information of images without modifying the pre-trained ViT structure, in order to enhance the performance of image aesthetic assessment and image quality assessment.

Chat-based Person Retrieval via Dialogue-Refined Cross-Modal Alignment

Yang Bai (Wuhan University), Mang Ye (Wuhan University)

CodeRetrievalTransformerLarge Language ModelContrastive LearningImageTextMultimodality

🎯 What it does: A framework for dialogue-oriented person retrieval (ChatPR) is proposed and implemented, which refines the retrieval query through interactive multi-turn dialogue and constructs the first ChatPedes dataset using dialogue data generated by large language models;

Cheb-GR: Rethinking K-nearest Neighbor Search in Re-ranking for Person Re-identification

Jinxi Yang (Wuhan University), Mang Ye (Wuhan University)

CodeRecognitionRetrievalGraph Neural NetworkImage

🎯 What it does: The Cheb-GR method is proposed, utilizing adaptive neighbor search guided by the Chebyshev theorem and graph convolution rearrangement, eliminating the reliance on k-nearest neighbor search in traditional rearrangement.

CheXwhatsApp: A Dataset for Exploring Challenges in the Diagnosis of Chest X-rays through Mobile Devices

Mariamma Antony (Indian Institute of Science), Chiranjib Bhattacharyya (Indian Institute of Science)

CodeClassificationObject DetectionCompressionExplainability and InterpretabilityConvolutional Neural NetworkImage

🎯 What it does: The CheXwhatsApp dataset has been constructed and made publicly available, containing 141,804 pairs of original and WhatsApp-compressed chest X-ray images. Three quantification metrics, PIP, OLS, and LI, are proposed to evaluate the prediction, interpretability, and localization instability caused by compression.

CheXWorld: Exploring Image World Modeling for Radiograph Representation Learning

Yang Yue (Tsinghua University), Gao Huang (Tsinghua University)

CodeClassificationSegmentationRepresentation LearningTransformerContrastive LearningWorld ModelImageBiomedical Data

🎯 What it does: This paper proposes CheXWorld, a self-supervised world modeling framework for chest X-ray images, aimed at learning anatomical structures, layouts, and domain transformation knowledge.

CholecTrack20: A Multi-Perspective Tracking Dataset for Surgical Tools

Chinedu Innocent Nwoye (University of Strasbourg), Nicolas Padoy (University of Strasbourg)

CodeObject DetectionObject TrackingVideoBenchmark

🎯 What it does: Created and publicly released the CholecTrack20 multi-view surgical tool tracking dataset, which includes multi-class and multi-tool tracking annotations for 20 complete cholecystectomy surgery videos.

Circumventing Shortcuts in Audio-visual Deepfake Detection Datasets with Unsupervised Learning

Stefan Smeu (Bitdefender), Elisabeta Oneata (Bitdefender)

CodeClassificationAnomaly DetectionContrastive LearningVideoMultimodalityAudio

🎯 What it does: This paper reveals the existence of a 'leading silence' short-term silence bias in audio-video deepfake datasets and proposes an unsupervised learning framework called AVH-Align, which utilizes only real samples. It enhances detection robustness by aligning AV-HuBERT self-supervised features, avoiding reliance on dataset shortcuts like brief silences.

Classifier-Free Guidance Inside the Attraction Basin May Cause Memorization

Anubhav Jain (New York University), Yuki Mitsufuji (Sony Group Corporation)

CodeGenerationDiffusion modelImage

🎯 What it does: This study investigates the phenomenon of memorization in diffusion models and proposes techniques to reduce memorization by detecting and avoiding attractor basins in the diffusion trajectory, using dynamic or static transition points and Opposite Guidance.

Classifier-guided CLIP Distillation for Unsupervised Multi-label Classification

Dongseob Kim (Samsung Electronics), Hyunjung Shim (KAIST)

CodeClassificationKnowledge DistillationContrastive LearningImage

🎯 What it does: This paper proposes an unsupervised multi-label classification method called Classifier-guided CLIP Distillation (CCD), which utilizes pseudo-labels from CLIP and combines classifier-guided local views and debiasing techniques to train multi-label recognition models without manual annotation.

Classifier-to-Bias: Toward Unsupervised Automatic Bias Detection for Visual Classifiers

Quentin Guimard (University of Trento), Elisa Ricci (University of Trento)

CodeClassificationRetrievalLarge Language ModelImageTextRetrieval-Augmented Generation

🎯 What it does: An unsupervised framework named CLASSIFIER-TO-BIAS (C2B) is proposed, which can automatically identify the biases of visual classifiers solely based on the textual description of the given classification task and a pre-trained model.

ClearSight: Visual Signal Enhancement for Object Hallucination Mitigation in Multimodal Large Language Models

Hao Yin (University of Science and Technology of China), Zilei Wang (University of Science and Technology of China)

CodeObject DetectionGenerationTransformerLarge Language ModelVision Language ModelImageTextMultimodality

🎯 What it does: A training-free and tool-free visual enhancement fusion method (VAF) is proposed, which enhances the attention to visual signals in multimodal large language models by adjusting the mid-layer attention weights, thereby reducing the generation of false objects.

CLIP-driven Coarse-to-fine Semantic Guidance for Fine-grained Open-set Semi-supervised Learning

Xiaokun Li (Beijing Jiaotong University), Qingji Guan (Beijing Jiaotong University)

CodeClassificationRecognitionTransformerVision Language ModelContrastive LearningImage

🎯 What it does: A coarse-fine layered semantic guidance framework based on CLIP (CFSG-CLIP) is designed, which first filters local visual features related to categories through a semantic filtering module, and then further refines the features in the fine-grained branch through a visual semantic injection strategy to achieve efficient differentiation between fine-grained IDs and OOD samples.

CLOC: Contrastive Learning for Ordinal Classification with Multi-Margin N-pair Loss

Dileepa Pitawela (University of Adelaide), Hsiang-Ting Chen (University of Adelaide)

CodeClassificationExplainability and InterpretabilityConvolutional Neural NetworkContrastive LearningImage

🎯 What it does: This paper proposes an ordinal classification method CLOC based on multi-margin N-pair contrastive learning, which utilizes learnable margins to preserve category order and reduce critical decision errors.

Co-Speech Gesture Video Generation with Implicit Motion-Audio Entanglement

Xinjie Li (Pennsylvania State University), Jing Xiao (PingAn Technology)

CodeGenerationData SynthesisPose EstimationTransformerDiffusion modelVideoAudio

🎯 What it does: Proposes the Implicit Motion-Audio Entanglement (IMAE) method, which generates realistic upper-body pose videos directly from audio without using additional inputs.

CO-SPY: Combining Semantic and Pixel Features to Detect Synthetic Images by AI

Siyuan Cheng (Purdue University), Vikash Sehwag (Sony AI)

CodeData SynthesisAnomaly DetectionAuto EncoderContrastive LearningImageBenchmark

🎯 What it does: Proposes the CO-Spy framework, which combines semantic features and pixel-level traces to detect AI-generated images.

CoA: Towards Real Image Dehazing via Compression-and-Adaptation

Long Ma, Zhuo Su

CodeRestorationDomain AdaptationOptimizationComputational EfficiencyKnowledge DistillationContrastive LearningImage

🎯 What it does: A dual-stage learning framework based on compression and adaptation (CoA) is proposed, which performs adaptation through dual-layer optimization in the real domain after compressing the model in the synthetic domain;

COB-GS: Clear Object Boundaries in 3DGS Segmentation Based on Boundary-Adaptive Gaussian Splitting

Jiaxin Zhang (Harbin Institute of Technology), Xianming Liu (Harbin Institute of Technology)

CodeSegmentationOptimizationGaussian SplattingPoint Cloud

🎯 What it does: This paper addresses the issue of boundary ambiguity in 3D Gaussian Splatting (3DGS) for object segmentation and proposes a joint optimization method for semantics and texture segmentationβ€”COB-GS.

CocoER: Aligning Multi-Level Feature by Competition and Coordination for Emotion Recognition

Xuli Shen (Fudan University), Xiangyang Xue (Fudan University)

CodeRecognitionTransformerContrastive LearningImageMultimodality

🎯 What it does: A multi-layer visual feature alignment framework called CocoER is proposed, which is based on competitive and collaborative mechanisms. It extracts head, body, and background features through cross-layer attention and generates pseudo-labels via a vocabulary information alignment module to guide feature refinement.

CoE: Chain-of-Explanation via Automatic Visual Concept Circuit Description and Polysemanticity Quantification

Wenlong Yu (Tianjin University), Qinghua Hu (Tianjin University)

CodeExplainability and InterpretabilityConvolutional Neural NetworkTransformerLarge Language ModelVision Language ModelImage

🎯 What it does: A deep visual model interpretability framework based on automated concept decoding, separation, ambiguity quantification, and chain-based explanation is proposed.

Collaborative Decoding Makes Visual Auto-Regressive Modeling Efficient

Zigeng Chen (National University of Singapore), Xinchao Wang (National University of Singapore)

CodeGenerationComputational EfficiencyKnowledge DistillationTransformerSupervised Fine-TuningImage

🎯 What it does: This paper proposes Collaborative Decoding (CoDe), which splits the coarse and fine scale generation tasks of visual autoregressive models into a large model responsible for low-frequency global sketches and a small model responsible for high-frequency details, significantly improving inference speed and memory utilization.

Color Alignment in Diffusion

Ka Chun Shum (Hong Kong University of Science and Technology), Sai-Kit Yeung (Hong Kong University of Science and Technology)

CodeGenerationData SynthesisDiffusion modelImage

🎯 What it does: This paper proposes a method for achieving fine-grained color conditional generation in diffusion models, capable of generating images that are highly consistent with given color patterns (color values and proportions) while maintaining diversity and quality of content.

ComfyBench: Benchmarking LLM-based Agents in ComfyUI for Autonomously Designing Collaborative AI Systems

Xiangyuan Xue (Shanghai Artificial Intelligence Laboratory), Lei Bai (Shanghai Artificial Intelligence Laboratory)

CodeLarge Language ModelAgentic AIBenchmarkRetrieval-Augmented Generation

🎯 What it does: A ComfyBench benchmark was constructed, and ComfyAgent was proposed to automatically design and generate ComfyUI workflows using LLM.

CoMM: A Coherent Interleaved Image-Text Dataset for Multimodal Understanding and Generation

Wei Chen (Hong Kong University of Science and Technology), Long Chen (Hong Kong University of Science and Technology)

CodeGenerationData SynthesisTransformerLarge Language ModelReinforcement LearningVision Language ModelImageTextMultimodalityBenchmark

🎯 What it does: We constructed and released CoMMβ€”a high-quality, coherent interleaved text-image datasetβ€”and proposed four new benchmarks for interleaved text-image generation and evaluation.

Complementary Advantages: Exploiting Cross-Field Frequency Correlation for NIR-Assisted Image Denoising

Yuchen Wang (Beijing Institute of Technology), Hua Huang (Beijing Normal University)

CodeRestorationConvolutional Neural NetworkImage

🎯 What it does: This paper proposes an image denoising network called FCENet, which utilizes near-infrared (NIR) images to assist in achieving high-quality denoising of RGB images through frequency domain information fusion.

Compositional Targeted Multi-Label Universal Perturbations

Hassan Mahmood (Northeastern University), Ehsan Elhamifar (Northeastern University)

CodeOptimizationAdversarial AttackImage

🎯 What it does: A combinatorial-based multi-label target universal perturbation generation framework is proposed, capable of synthesizing attack perturbations for any label combination in linear time;

ComRoPE: Scalable and Robust Rotary Position Embedding Parameterized by Trainable Commuting Angle Matrices

Hao Yu (Tsinghua University), Chun Yuan (Shenzhen University)

CodeClassificationObject DetectionTransformerImage

🎯 What it does: A trainable and exchangeable angle matrix rotation position encoding (ComRoPE) is proposed, achieving scalability and robustness for Transformer position encoding.

Conformal Prediction for Zero-Shot Models

Julio Silva-RodrΓ­guez (Γ‰cole de Technologie SupΓ©rieure), Jose Dolz (Γ‰cole de Technologie SupΓ©rieure)

CodeClassificationDomain AdaptationTransformerVision Language ModelImageMultimodality

🎯 What it does: This study investigates how to apply split conformal prediction to large-scale pre-trained vision-language models (such as CLIP) and proposes an unsupervised transfer learning framework based on optimal transport, named Conf-OT, to address the distribution shift between the source and target domains.

Consistency Posterior Sampling for Diverse Image Synthesis

Vishal Purohit (Purdue University), Xiuyuan Cheng (Duke University)

CodeRestorationGenerationData SynthesisSuper ResolutionDiffusion modelImageStochastic Differential Equation

🎯 What it does: A posterior sampling method based on Langevin dynamics in the noise space of pre-trained generative models is proposed, which can efficiently generate diverse posterior samples.

Context-Aware Multimodal Pretraining

Karsten Roth (Tuebingen AI Center), Olivier J. Henaff (Google DeepMind)

CodeRepresentation LearningTransformerContrastive LearningImageMultimodality

🎯 What it does: This paper proposes Context-Aware Multimodal Pretraining (LIxP), which incorporates cross-attention contextual buffering into contrastive pretraining, enabling visual-text models to perform metric-based few-shot adaptation without additional training after training.

ConText-CIR: Learning from Concepts in Text for Composed Image Retrieval

Eric Xing (Washington University in St. Louis), Nathan Jacobs (Washington University in St. Louis)

CodeRetrievalTransformerContrastive LearningImageText

🎯 What it does: The ConText-CIR framework is proposed, which trains with Text Concept-Consistency loss and achieves efficient image-text combined retrieval through cross-modal attention.

Context-Enhanced Memory-Refined Transformer for Online Action Detection

Zhanzhong Pang (National University of Singapore), Angela Yao (National University of Singapore)

CodeRecognitionTransformerVideo

🎯 What it does: To address the frame representation learning bias caused by the inconsistency between training and inference in Online Action Detection (OAD), a new Transformer framework called Context-Enhanced Memory-Refined Transformer (CMeRT) is proposed.

Contextual AD Narration with Interleaved Multimodal Sequence

Hanlin Wang (Nanjing University), Limin Wang (Nanjing University)

CodeRecognitionGenerationTransformerLarge Language ModelContrastive LearningVideoTextMultimodalityAudio

🎯 What it does: A unified framework called Uni-AD is proposed, which automatically generates audio descriptions (AD) using interleaved multimodal sequences and pre-trained LLMs.

Continual SFT Matches Multimodal RLHF with Negative Supervision

Ke Zhu (Nanjing University), Jingdong Wang (Baidu)

CodeComputational EfficiencyReinforcement Learning from Human FeedbackTransformerLarge Language ModelSupervised Fine-TuningReinforcement LearningVision Language ModelImageTextMultimodality

🎯 What it does: Proposes a negative supervision fine-tuning (nSFT) method that transforms negative feedback from RLHF into SFT loss using a single model, thereby efficiently aligning visual language models.

Continuous Locomotive Crowd Behavior Generation

Inhwan Bae (Gwangju Institute of Science and Technology), Hae-Gon Jeon (Gwangju Institute of Science and Technology)

CodeGenerationData SynthesisTransformerDiffusion modelImageVideo

🎯 What it does: A continuous crowd behavior generation framework called CrowdES is designed based on diffusion models and state-switching dynamics, capable of automatically generating diverse and realistic crowd trajectories continuously from a single scene image while supporting user interactive control.

Correcting Deviations from Normality: A Reformulated Diffusion Model for Multi-Class Unsupervised Anomaly Detection

Farzad Beizaee (Γ‰cole Nationale SupΓ©rieure de l'Γ‰lectronique et de ses Applications), Jose Dolz (Γ‰cole Nationale SupΓ©rieure de l'Γ‰lectronique et de ses Applications)

CodeAnomaly DetectionDiffusion modelAuto EncoderImage

🎯 What it does: A rewritten diffusion model DeCo-Diff is proposed, which only corrects abnormal regions while keeping normal areas unchanged in multi-class unsupervised anomaly detection.

CoSDH: Communication-Efficient Collaborative Perception via Supply-Demand Awareness and Intermediate-Late Hybridization

Junhao Xu (Beihang University), Di Huang (Beihang University)

CodeObject DetectionAutonomous DrivingAuto EncoderPoint Cloud

🎯 What it does: The CoSDH framework is proposed, utilizing supply-demand awareness and late-stage mixed fusion to achieve efficient communication for collaborative 3D object detection.

COSMOS: Cross-Modality Self-Distillation for Vision Language Pre-training

Sanghwan Kim (Technical University of Munich), Zeynep Akata (Technical University of Munich)

CodeClassificationSegmentationRetrievalKnowledge DistillationTransformerVision Language ModelContrastive LearningTextMultimodality

🎯 What it does: This paper proposes COSMOS, which utilizes cross-modal self-distillation for pre-training visual language models;

Crab: A Unified Audio-Visual Scene Understanding Model with Explicit Cooperation

Henghui Du (Renmin University of China), Di Hu (Renmin University of China)

CodeRecognitionObject DetectionSegmentationData-Centric LearningTransformerLarge Language ModelSupervised Fine-TuningVision Language ModelVideoMultimodalityAudio

🎯 What it does: A unified audio-video scene understanding model, Crab, is proposed, capable of simultaneously working on various audio-video tasks such as temporal localization, spatial localization, spatiotemporal reasoning, and pixel-level understanding.

Critic-V: VLM Critics Help Catch VLM Errors in Multimodal Reasoning

Di Zhang (Fudan University), Dongzhan Zhou (Shanghai Artificial Intelligence Laboratory)

CodeOptimizationReinforcement Learning from Human FeedbackTransformerLarge Language ModelReinforcement LearningVision Language ModelTextMultimodality

🎯 What it does: The Critic-V framework is proposed, which splits the reasoning process of visual language models into a Reasoner and an external Critic, iteratively improving the reasoning path through natural language criticism.

CroCoDL: Cross-device Collaborative Dataset for Localization

Hermann Blum (Lamarr Institute for AI and Security), Zuria Bauer (ETH Zurich)

CodePose EstimationRetrievalRobotic IntelligenceSimultaneous Localization and MappingMultimodalityBenchmark

🎯 What it does: The CroCoDL dataset has been constructed and released, providing multimodal visual localization data recorded across 10 different scenes and 4 types of devices (robots, handheld smartphones, MR headsets, NavVis scanners), and generating high-precision pseudo Ground-Truth based on LaMAR.

Cross-modal Causal Relation Alignment for Video Question Grounding

Weixing Chen (Sun Yat-sen University), Liang Lin (Sun Yat-sen University)

CodeRecognitionRetrievalOptimizationTransformerContrastive LearningGaussian SplattingVideoMultimodality

🎯 What it does: A cross-modal causal relationship alignment framework (CRA) is proposed, utilizing Gaussian Smoothing Grounding, Cross-Modal Alignment, and Explicit Causal Intervention to achieve weakly supervised video question answering and temporal localization.

Cross-Modal Distillation for 2D/3D Multi-Object Discovery from 2D Motion

Saad Lahlali (Universite Paris-Saclay), Quoc-Cuong Pham (Universite Paris-Saclay)

CodeObject DetectionAutonomous DrivingKnowledge DistillationOptical FlowMultimodalityPoint Cloud

🎯 What it does: This paper proposes a 3D multi-object discovery method DIOD-3D based on 2D motion information, and achieves mutual supervision between 2D and 3D through the cross-modal distillation framework xMOD, enabling unsupervised multi-modal object discovery.

Cross-Modal Interactive Perception Network with Mamba for Lung Tumor Segmentation in PET-CT Images

Jie Mei (Hunan University), Dong Dai (Tianjin Medical University Cancer Institute and Hospital)

CodeSegmentationImageBiomedical DataComputed TomographyPositron Emission Tomography

🎯 What it does: This study first collected and annotated 21,930 pairs of PET-CT lung images, constructing the publicly available dataset PCLT20K; subsequently, a cross-modal interactive perception network (CIPA) based on the Mamba state space model was proposed for lung tumor segmentation in PET-CT images.

Cross-Rejective Open-Set SAR Image Registration

Shasha Mao (Xidian University), Yimeng Zhang (Xidian University)

CodeClassificationRecognitionTransformerContrastive LearningImage

🎯 What it does: To address the problem of SAR image registration, a Cross Reject Open Set Registration method (CroR-OSIR) is proposed, which achieves accurate selection and localization of matching points through cross open set classification and supervised contrastive learning in two image domains.

CSC-PA: Cross-image Semantic Correlation via Prototype Attentions for Single-network Semi-supervised Breast Tumor Segmentation

Zhenhui Ding (Shenzhen University), Jing Qin (Hong Kong Polytechnic University)

CodeSegmentationConvolutional Neural NetworkImageUltrasound

🎯 What it does: A single-network semi-supervised breast ultrasound image segmentation framework CSC-PA is proposed, which enhances segmentation performance by utilizing cross-image semantic correlation.

Cubify Anything: Scaling Indoor 3D Object Detection

Justin Lazarow (Apple), Afshin Dehghan (Apple)

CodeObject DetectionTransformerPoint Cloud

🎯 What it does: A large-scale indoor 3D object detection dataset CA-1M is proposed, along with the CuTR model based on Transformer.

Curriculum Coarse-to-Fine Selection for High-IPC Dataset Distillation

Yanda Chen (Harbin Institute of Technology), Liqiang Nie (Harbin Institute of Technology)

CodeKnowledge DistillationImage

🎯 What it does: This paper proposes a course-based coarse-to-fine selection method named CCFS for ensemble data distillation under high per-class sample (high-IPC) conditions, enhancing the distillation effect by gradually selecting suitable real images combined with synthetic images.

CustomKD: Customizing Large Vision Foundation for Edge Model Improvement via Knowledge Distillation

Jungsoo Lee (Qualcomm AI Research), Fatih Porikli (Qualcomm AI Research)

CodeDomain AdaptationKnowledge DistillationConvolutional Neural NetworkTransformerImage

🎯 What it does: A new knowledge distillation framework called CustomKD is proposed, utilizing large-scale visual foundation models (such as DINOv2 and OpenCLIP) as teachers. Through alternating training of feature customization and distillation, it significantly enhances the performance of edge models (such as MobileNetV3 and WideResNet) in scenarios with unlabeled data (Unsupervised Domain Adaptation UDA and Semi-Supervised Learning SSL).

D^3CTTA: Domain-Dependent Decorrelation for Continual Test-Time Adaption of 3D LiDAR Segmentation

Jichun Zhao (University of Chinese Academy of Sciences), Dong Gong (University of New South Wales)

CodeSegmentationDomain AdaptationAutonomous DrivingPoint Cloud

🎯 What it does: This paper proposes a Continuous Testing Time Adaptation (CTTA) framework DCTTA, designed for LiDAR semantic segmentation, which can online update the model in continuously changing environments.

DA-VPT: Semantic-Guided Visual Prompt Tuning for Vision Transformers

Li Ren (University of Central Florida), Kien Hua (University of Central Florida)

CodeClassificationSegmentationTransformerPrompt EngineeringContrastive LearningImage

🎯 What it does: To achieve parameter-efficient fine-tuning of Vision Transformers, we propose Distribution-Aware Visual Prompt Tuning (DA-VPT).

DAMM-Diffusion: Learning Divergence-Aware Multi-Modal Diffusion Model for Nanoparticles Distribution Prediction

Junjie Zhou (Nanjing University of Aeronautics and Astronautics), Wei Shao (Nanjing University of Aeronautics and Astronautics)

CodeGenerationData SynthesisDiffusion modelMultimodalityBiomedical Data

🎯 What it does: A diffusion model called DAMM-Diffusion, which integrates single-modal and multi-modal branches, is proposed for predicting the distribution of nanoparticles under tumor microenvironment conditions (vascular and cell nuclei).

DarkIR: Robust Low-Light Image Restoration

Daniel Feijoo (Cidaut AI), Marcos V. Conde (Computer Vision Lab)

CodeRestorationConvolutional Neural NetworkImage

🎯 What it does: A lightweight multi-task network called DarkIR is proposed for simultaneous denoising, deblurring, and low-light enhancement in nighttime/low-light environments.

Data Distributional Properties As Inductive Bias for Systematic Generalization

Felipe del Rio (Pontificia Universidad Catolica de Chile), Alvaro Soto (Pontificia Universidad Catolica de Chile)

CodeTransformerMultimodality

🎯 What it does: Exploring how the characteristics of training data distribution (diversity, burstiness, latent intervention) enhance the system generalization of multimodal models.

Data Synthesis with Diverse Styles for Face Recognition via 3DMM-Guided Diffusion

Yuxi Mi (Fudan University), Shuigeng Zhou (Tencent)

CodeRecognitionGenerationData SynthesisDiffusion modelImage

🎯 What it does: This paper proposes a diffusion model named MorphFace, which can generate identity-preserving and stylistically diverse synthetic faces to enhance the diversity of training data for facial recognition models.

Debiasing Multimodal Large Language Models via Noise-Aware Preference Optimization

Zefeng Zhang (Institute of Information Engineering, Chinese Academy of Sciences), Tingwen Liu (Institute of Information Engineering, Chinese Academy of Sciences)

CodeOptimizationTransformerLarge Language ModelReinforcement LearningMultimodalityBenchmark

🎯 What it does: This paper proposes a multimodal large language model debiasing method based on preference optimization, first constructing the RLAIF-V-Bias dataset and designing the NaPO algorithm to suppress modal bias.

DeCLIP: Decoupled Learning for Open-Vocabulary Dense Perception

Junjie Wang (Harbin Institute of Technology), Zhuotao Tian (Harbin Institute of Technology)

CodeObject DetectionSegmentationKnowledge DistillationRepresentation LearningTransformerContrastive LearningImage

🎯 What it does: The DeCLIP framework is proposed, which decouples the self-attention module of CLIP to separately learn local identification ability and spatial consistency, enhancing the dense prediction performance of open vocabulary.

Decoder Gradient Shield: Provable and High-Fidelity Prevention of Gradient-Based Box-Free Watermark Removal

Haonan An (City University of Hong Kong), Yuguang Fang (City University of Hong Kong)

CodeImage TranslationAdversarial AttackConvolutional Neural NetworkGenerative Adversarial NetworkImage

🎯 What it does: This paper addresses a black-box watermarking method for image-to-image models, revealing the vulnerability of the decoder to gradient attacks, and proposes the Decoder Gradient Shield (DGS) mechanism, which redirects and scales gradients in a black-box API to prevent attackers from training a de-watermarking network.

Decoupling Fine Detail and Global Geometry for Compressed Depth Map Super-Resolution

Huan Zheng, Jianbing Shen

CodeRestorationDepth EstimationSuper ResolutionConvolutional Neural NetworkImage

🎯 What it does: A geometric decoupling network named GDNet is proposed for recovering high-quality depth maps from compressed depth images.

DeDe: Detecting Backdoor Samples for SSL Encoders via Decoders

Sizai Hou (Hong Kong University of Science and Technology), Duanyi Yao (Hong Kong University of Science and Technology)

CodeAnomaly DetectionRepresentation LearningConvolutional Neural NetworkContrastive LearningImage

🎯 What it does: A decoder-based reverse mapping method called DeDe is proposed for detecting hidden backdoor samples in unsupervised self-supervised learning (SSL) encoders.

Deep Change Monitoring: A Hyperbolic Representative Learning Framework and a Dataset for Long-term Fine-grained Tree Change Detection

Yante Li, Guoying Zhao

CodeRepresentation LearningConvolutional Neural NetworkImage

🎯 What it does: A UAVTC dataset for long-term fine-grained tree change detection is proposed, and a hypercurvature Siamese network (HSN) is designed to capture the hierarchical structure of tree changes.

DeepCompress-ViT: Rethinking Model Compression to Enhance Efficiency of Vision Transformers at the Edge

Sabbir Ahmed (Binghamton University), Adnan Siraj Rakin (Binghamton University)

CodeClassificationObject DetectionCompressionComputational EfficiencyKnowledge DistillationTransformerAuto EncoderImage

🎯 What it does: Designed and implemented DeepCompress-ViT, which significantly compresses the Vision Transformer using an encoder-decoder structure, and ensures accuracy is not lost through optimized inference-time decoding.

DeformCL: Learning Deformable Centerline Representation for Vessel Extraction in 3D Medical Image

Ziwei Zhao (Yizhun Medical AI Co), Liwei Wang (Center for Machine Learning Research, Peking University)

CodeSegmentationTransformerBiomedical DataMagnetic Resonance Imaging

🎯 What it does: A continuous representation based on Deformable Centerline (DeformCL) is proposed for the extraction and segmentation of blood vessels in 3D medical images;

Degradation-Aware Feature Perturbation for All-in-One Image Restoration

Xiangpeng Tian (Sichuan University), Chao Ren (Sichuan University)

CodeRestorationTransformerPrompt EngineeringImage

🎯 What it does: A global unified model DFPIR based on degradation-aware feature perturbation is designed, capable of handling multiple degradation tasks such as image denoising, dehazing, deraining, deblurring, and low-light enhancement in one go.

DejaVid: Encoder-Agnostic Learned Temporal Matching for Video Classification

Darryl Ho (Massachusetts Institute of Technology), Samuel Madden (Massachusetts Institute of Technology)

CodeClassificationTransformerVideo

🎯 What it does: An Encoder-agnostic DejaVid method is proposed, which encodes videos into variable-length time series embeddings (TSE) and uses Dynamic Time Warping (DTW) with learnable temporal feature weights to enhance video classification performance.

DELT: A Simple Diversity-driven EarlyLate Training for Dataset Distillation

Zhiqiang Shen (MBZUAI), Shitong Shao (MBZUAI)

CodeData SynthesisComputational EfficiencyKnowledge DistillationImage

🎯 What it does: Proposes the EarlyLate training strategy, which optimizes synthetic samples in stages to enhance the diversity and efficiency of dataset distillation under batch-to-global matching.

Depth-Guided Bundle Sampling for Efficient Generalizable Neural Radiance Field Reconstruction

Li Fang (Communication University of China), Zhan Ma (Nanjing University)

CodeData SynthesisComputational EfficiencyNeural Radiance FieldPoint Cloud

🎯 What it does: A depth-guided beam sampling strategy is proposed, which groups adjacent rays for joint sampling and adaptively controls the number of sampling points to improve the rendering speed and quality of general NeRF.

Derivative-Free Diffusion Manifold-Constrained Gradient for Unified XAI

Won Jun Kim (KAIST), Jong Chul Ye (KAIST)

CodeExplainability and InterpretabilityDiffusion modelImageBiomedical Data

🎯 What it does: This paper proposes a FreeMCG method based on diffusion models and EnKF for the unified implementation of black-box feature importance explanation and counterfactual explanation.

DeSiRe-GS: 4D Street Gaussians for Static-Dynamic Decomposition and Surface Reconstruction for Urban Driving Scenes

Chensheng Peng (University of California Berkeley), Wei Zhan (University of California Berkeley)

CodeAutonomous DrivingGaussian SplattingPoint Cloud

🎯 What it does: This paper proposes DeSiRe-GS, a self-supervised static-dynamic decomposition and surface reconstruction framework based on 4D Gaussian splatting.

Detect Any Mirrors: Boosting Learning Reliability on Large-Scale Unlabeled Data with an Iterative Data Engine

Zhaohu Xing (Hong Kong University of Science and Technology), Lei Zhu (Hong Kong University of Science and Technology)

CodeObject DetectionSegmentationAnomaly DetectionKnowledge DistillationTransformerLarge Language ModelImageVideoMultimodality

🎯 What it does: By constructing a semi-supervised mirror detection framework and collecting approximately 400,000 unlabeled mirror images, the quality of pseudo-labels is improved using an iterative data engine, achieving more robust mirror detection.

Detect-and-Guide: Self-regulation of Diffusion Models for Safe Text-to-Image Generation via Guideline Token Optimization

Feifei Li (Fudan University), Min Yang (Fudan University)

CodeGenerationDiffusion modelImageText

🎯 What it does: This paper proposes the Detect-and-Guide (DAG) framework, which performs self-inspection in text-to-image diffusion models by optimizing prompt words and extracting Class Activation Maps (CAM). It then employs adaptive safety guidance during the sampling process to finely eliminate detected unsafe areas, achieving safe generation of sexual content.

Detecting Adversarial Data Using Perturbation Forgery

Qian Wang (Huazhong University of Science and Technology), Ning Yu (Netflix Eyeline)

CodeAnomaly DetectionAdversarial AttackDiffusion modelGenerative Adversarial NetworkImage

🎯 What it does: Proposes the Perturbation Forgery method, which perturbs the noise distribution of known attacks, generates sparse masks, and synthesizes pseudo-adversarial samples to train a binary classifier for detecting unknown adversarial attacks.

Detecting Backdoor Attacks in Federated Learning via Direction Alignment Inspection

Jiahao Xu (University of Nevada), Rui Hu (University of Nevada)

CodeAnomaly DetectionFederated LearningImage

🎯 What it does: A backdoor defense method based on directional alignment detection, AlignIns, is proposed to identify and filter malicious model updates in federated learning.

Detecting Open World Objects via Partial Attribute Assignment

Muli Yang (Institute for Infocomm Research), Hongyuan Zhu (Institute for Infocomm Research)

CodeObject DetectionContrastive LearningImageBenchmark

🎯 What it does: This paper proposes the PASS method, which automatically selects and optimizes a small subset of attributes from a large attribute pool through partial optimal transport and curriculum learning, to achieve simultaneous detection of known and unknown objects in open-world object detection (OWOD).

Detecting Out-of-Distribution Through the Lens of Neural Collapse

Litian Liu (Massachusetts Institute of Technology), Yao Qin (University of California Santa Barbara)

CodeClassificationAnomaly DetectionConvolutional Neural NetworkTransformerImage

🎯 What it does: This paper proposes a post-hoc OOD detection method based on the theory of Neural Collapse, utilizing the alignment of ID sample features with the last layer weight vectors and the L1 norm of the features to distinguish OOD samples.

Detection-Friendly Nonuniformity Correction: A Union Framework for Infrared UAV Target Detection

Houzhang Fang (Xidian University), Luxin Yan (Huazhong University of Science and Technology)

CodeRestorationObject DetectionTransformerImageBenchmark

🎯 What it does: A joint, detection-friendly end-to-end framework called UniCD is proposed for simultaneously performing non-uniformity correction and target detection of infrared drone images under low-frequency non-uniformity conditions.

Deterministic Image-to-Image Translation via Denoising Brownian Bridge Models with Dual Approximators

Bohan Xiao (Wayne State University), Ming Dong (Wayne State University)

CodeImage TranslationGenerationScore-based ModelImageStochastic Differential Equation

🎯 What it does: A dual-approximation denoising Brownian bridge model (Dual‑approx Bridge) is proposed, achieving deterministic image-to-image translation.

Devils in Middle Layers of Large Vision-Language Models: Interpreting, Detecting and Mitigating Object Hallucinations via Attention Lens

Zhangqi Jiang (National University of Defense Technology), Xu Yang (Southeast University)

CodeObject DetectionGenerationExplainability and InterpretabilityTransformerVision Language ModelImage

🎯 What it does: This paper conducts an in-depth study of the object hallucination mechanism in large visual-language models (LVLM) from the perspective of attention and proposes a no-training-cost hallucination suppression method based on intermediate layer attention modulation.

DFormerv2: Geometry Self-Attention for RGBD Semantic Segmentation

Bo-Wen Yin (Nankai University), Qibin Hou (Nankai University)

CodeSegmentationTransformerMultimodality

🎯 What it does: DFormerv2 is proposed, a RGB-D semantic segmentation model that directly guides self-attention using the geometric priors of depth maps, without the need for an additional depth encoder, significantly improving segmentation performance.

DiC: Rethinking Conv3x3 Designs in Diffusion Models

Yuchuan Tian (Peking University), Hanting Chen (Huawei)

CodeGenerationData SynthesisConvolutional Neural NetworkDiffusion modelImage

🎯 What it does: A diffusion model DiC based on pure 3x3 convolutions is proposed, achieving image generation under the Encoder-Decoder hourglass architecture.

Diff-Palm: Realistic Palmprint Generation with Polynomial Creases and Intra-Class Variation Controllable Diffusion Models

Jianlong Jin (Hefei University of Technology), Yunsheng Wu (Tencent Youtu Lab)

CodeRecognitionGenerationData SynthesisDiffusion modelImage

🎯 What it does: A Diff-Palm model based on polynomial line drawing and K-step noise sharing sampling is proposed to generate high-quality, controllable variations of palm print images on a large scale, and to directly train recognition networks without fine-tuning on real data.

DiffLO: Semantic-Aware LiDAR Odometry with Diffusion-Based Refinement

Yongshu Huang (Xiamen University), Cheng Wang (Xiamen University)

CodePose EstimationAutonomous DrivingKnowledge DistillationDiffusion modelPoint Cloud

🎯 What it does: A LiDAR pose estimation network called DiffLO is proposed, which integrates semantic awareness and diffusion models to improve the accuracy and robustness of large-scale LiDAR pose estimation.

Diffusion-4K: Ultra-High-Resolution Image Synthesis with Latent Diffusion Models

Jinjin Zhang (Beihang University), Di Huang (Beihang University)

CodeGenerationData SynthesisDiffusion modelImageTextBenchmark

🎯 What it does: Proposed the Diffusion-4K framework, achieving direct generation of 4K level text-to-image, and constructed the Aesthetic-4K evaluation benchmark.

Diffusion-based Event Generation for High-Quality Image Deblurring

Xinan Xie (Sun Yat-sen University), Wei-Shi Zheng (Sun Yat-sen University)

CodeRestorationGenerationDiffusion modelImage

🎯 What it does: This paper proposes an event-guided image deblurring framework (EGDeblurring) based on a diffusion model, which can perform deblurring using only blurred images.

DiffusionSfM: Predicting Structure and Motion via Ray Origin and Endpoint Diffusion

Qitao Zhao (Carnegie Mellon University), Shubham Tulsiani (Carnegie Mellon University)

CodePose EstimationDepth EstimationTransformerDiffusion modelImage

🎯 What it does: An end-to-end multi-view method based on a denoising diffusion model is proposed, which directly predicts pixel-level ray origins and endpoints, thereby obtaining both scene geometry and camera poses simultaneously.

DiffVsgg: Diffusion-Driven Online Video Scene Graph Generation

Mu Chen (Zhejiang University), Yi Yang (Zhejiang University)

CodeObject DetectionGenerationDiffusion modelVideo

🎯 What it does: This paper proposes DIFFVSGG, a method for generating online video scene graphs based on latent diffusion models, which can update object categories, bounding boxes, and relationships in real-time for each frame, supporting continuous temporal reasoning.

DifIISR: A Diffusion Model with Gradient Guidance for Infrared Image Super-Resolution

Xingyuan Li (Dalian University of Technology), Jinyuan Liu (Dalian University of Technology)

CodeRestorationObject DetectionSegmentationSuper ResolutionDiffusion modelImage

🎯 What it does: This paper proposes a diffusion model-based infrared image super-resolution method called DifIISR, which achieves high-quality infrared image reconstruction by injecting visual and perceptual gradients during the reverse diffusion process.

DIFIX3D+: Improving 3D Reconstructions with Single-Step Diffusion Models

Jay Zhangjie Wu (NVIDIA), Huan Ling (NVIDIA)

CodeGenerationData SynthesisAutonomous DrivingDiffusion modelNeural Radiance FieldGaussian SplattingImage

🎯 What it does: This paper proposes a complete pipeline called DIFIX3D+ that utilizes a single-step diffusion model (DIFIX) to improve NeRF and 3D Gaussian Splatting for reconstruction and novel view synthesis.

DiGIT: Multi-Dilated Gated Encoder and Central-Adjacent Region Integrated Decoder for Temporal Action Detection Transformer

Ho-Joong Kim (Korea University), Seong-Whan Lee (Korea University)

CodeRecognitionObject DetectionTransformerVideo

🎯 What it does: A time action detection framework based on Transformer, called DiGIT, is proposed, which includes a multi-dilated gated encoder and a center-adjacent region integrated decoder.

Digital Twin Catalog: A Large-Scale Photorealistic 3D Object Digital Twin Dataset

Zhao Dong (Meta Reality Labs Research), Richard Newcombe (Meta Reality Labs Research)

CodeData SynthesisRobotic IntelligenceNeural Radiance FieldImagePoint CloudMesh

🎯 What it does: A Digital Twin Catalog (DTC) dataset has been proposed and released, containing high-precision 3D digital twin models of 2000 real objects, with millimeter-level geometric accuracy, photorealistic PBR materials, and multi-view evaluation data collected using DSLR and AR glasses.

DiN: Diffusion Model for Robust Medical VQA with Semantic Noisy Labels

Erjian Guo (University of Sydney), Luping Zhou (University of Sydney)

CodeTransformerDiffusion modelBiomedical Data

🎯 What it does: Proposes a noise label robust framework DiN for medical visual question answering.

Directional Label Diffusion Model for Learning from Noisy Labels

Senyu Hou (Shanxi University), Wenjian Wang (Shanxi University)

CodeClassificationDiffusion modelContrastive LearningImage

🎯 What it does: Designed and trained a dual-channel directional label diffusion model to learn high-quality classifications from noisy labels.

Disco4D: Disentangled 4D Human Generation and Animation from a Single Image

Hui En Pang (Nanyang Technological University), Ziwei Liu (Nanyang Technological University)

CodeGenerationData SynthesisPose EstimationDiffusion modelGaussian SplattingImage

🎯 What it does: Disco4D generates animatable 4D human models from a single image and separates the body from clothing using Gaussian Splatting;

Dispider: Enabling Video LLMs with Active Real-Time Interaction via Disentangled Perception, Decision, and Reaction

Rui Qian (Chinese University of Hong Kong), Jiaqi Wang (Shanghai Innovation Institute)

CodeGenerationOptimizationTransformerLarge Language ModelVision Language ModelVideoTextMultimodality

🎯 What it does: Designed and implemented Dispider, a decoupled framework for real-time interactive video large language models, supporting continuous monitoring, timely decision-making, and asynchronous responses;

Distilling Long-tailed Datasets

Zhenghao Zhao (University of Illinois Chicago), Yan Yan (National University of Singapore)

CodeKnowledge DistillationData-Centric LearningImage

🎯 What it does: This paper proposes the Dataset Distillation for Long-Tail Data (LTDD) task, which can compress imbalanced raw data into a small-scale, uniformly distributed synthetic dataset.