Scale-Localized Abstract Reasoning 抽象推理本地化规模 How Does Topology Influence Gradient Propagation and Model Performance of 拓扑如何影响梯度传播和模型性能 AdaBins Depth Estimation Using Adaptive Bins 使用自适应 bin 的 AdaBins 深度估计 Deep Burst Super-Resolution 超分辨率的深度爆发 Euro-PVI Pedestrian Vehicle Interactions in Dense Urban Centers 城市中心密集 Euro-PVI 行人和车辆互动 View Generalization for Single Image Textured 3D Models 查看单图像纹理 3D 模型的泛化 MetaHTR Towards Writer-Adaptive Handwritten Text Recognition MetaHTR 迈向作家自适应手写文本识别 More Photos Are All You Need Semi-Supervised Learning for Fine-Grained 更多的照片是你需要的所有细粒度的半监督学习 Vectorization and Rasterization Self-Supervised Learning for Sketch and Handwriting 草图和手写矢量化和光栅自我监督学习 Quantum Permutation Synchronization 量子置换同步 Behavior-Driven Synthesis of Human Dynamics 人类动力学的行为驱动综合 Understanding Object Dynamics for Interactive Image-to-Video Synthesis 从交互式图像到视频合成的对象动力学 Hardness Sampling for Self-Training Based Transductive Zero-Shot Learning 转导零样本学习的硬度采样是基于自我训练的 Hierarchical Video Prediction Using Relational Layouts for Human-Object Interactions 利用关系布局对人机交互进行分层视频预测 Convolutional Dynamic Alignment Networks for Interpretable Classifications 用于可解释和分类的卷积动态对齐网络 Towards Part-Based Understanding of RGB-D Scans 对 RGB-D 基于对部件的理解 InverseForm A Loss Function for Structured Boundary-Aware Segmentation 用于结构化边界感知分割 InverseForm A 损失函数 Few-Shot Segmentation Without Meta-Learning: A Good Transductive Inference Is All You Need?没有元学习的小样本分割一个好的转导推理就是一切 OCONet Image Extrapolation by Object Completion 通过对象完成进行 OCONet 图像外推 Neural Deformation Graphs for Globally-Consistent Non-Rigid Reconstruction 神经变形图用于全局一致非刚性重建 GAIA A Transfer Learning System of Object Detection That Fits GAIA 适适的目标检测迁移学习系统 Asymmetric Metric Learning for Knowledge Transfer 学习知识转移的不对称度 Fine-Grained Angular Contrastive Learning With Coarse Labels 学习粗标签的细粒度角度比较 Limitations of Post-Hoc Feature Alignment for Robustness 事后特征对齐鲁棒的限制 FBI-Denoiser Fast Blind Image Denoiser for Poisson-Gaussian Noise 泊松-高斯噪声用于泊松-高斯噪声 FBI-Denoiser 快速盲图像降噪器 Deep Lesion Tracker Monitoring Lesions in 4D Longitudinal Imaging Studies 监测深度病变跟踪器 4D 纵向成像研究中的病变 Exponential Moving Average Normalization for Self-Supervised and Semi-Supervised Learning 自我监督和半监督学习的指数移动平均归一化 Extreme Rotation Estimation Using Dense Correlation Volumes 估计使用密集相关体积的极端旋转 Rethinking Graph Neural Architecture Search From Message-Passing 从信息传递中重新思考图神经架构搜索 Revisiting Superpixels for Active Learning in Semantic Segmentation With Realistic 用真实的语义重新审视超像素，积极学习 Semantic Scene Completion via Integrating Instances and Scene In-the-Loop 语义场景通过集成实例和场景在环中完成 Debiased Subjective Assessment of Real-World Image Enhancement 真实世界图像增强的去偏主观评价 Normal Integration via Inverse Plane Fitting With Minimum Point-to-Plane Distance 正态积分通过具有最小点到平面距离的反平面拟合进行 ReMix Towards Image-to-Image Translation With Limited Data ReMix 以有限的数据转换图像 Sequential Graph Convolutional Network for Active Learning 主动学习的序列图卷积网络 MP3 A Unified Model To Map Perceive Predict and Plan MP3 统一的映射感知预测和计划模型 Architectural Adversarial Robustness The Case for Deep Pursuit 架构对抗鲁棒深度追求的案例 Deep Perceptual Preprocessing for Video Coding 预处理视频编码的深度感知 Ensembling With Deep Generative Views 集成深度生成视图 To the Point Efficient 3D Object Detection in the Range 范围内的点高效 3D 对象检测 Truly Shift-Invariant Convolutional Neural Networks 真正的移位不变卷积神经网络 BasicVSR The Search for Essential Components in Video Super-Resolution and BasicVSR 搜索视频超分辨率和基本组件 GLEAN Generative Latent Bank for Large-Factor Image Super-Resolution 用于大因子图像超分辨率 GLEAN Generative Latent Bank Pi-GAN Periodic Implicit Generative Adversarial Networks for 3D-Aware Image Synthesis 用于 3D 感知图像合成 Pi-GAN 周期性隐藏生成对抗网络 Adaptive Convolutions for Structure-Aware Style Transfer 结构感知风格迁移的自适应卷积 A Closer Look at Fourier Spectrum Discrepancies for CNN-Generated Images 仔细研究 CNN 傅立叶谱生成图像的差异 Learning Discriminative Prototypes With Dynamic Time Warping 学习有动态时间规则的判断原型 Towards Robust Classification Model by Counterfactual and Invariant Data Generation 通过反事实和不变数据生成稳定的分类模型 Your Flamingo is My Bird Fine-Grained or Not 你的火烈鸟是我的鸟 Conceptual 12M Pushing Web-Scale Image-Text Pre-Training To Recognize Long-Tail Visual 概念 12M 推送 Web 识别长尾视觉的大规模图像文本预训练 DexYCB A Benchmark for Capturing Hand Grasping of Objects DexYCB 用于捕获抓取物体的基准 On Focal Loss for Class-Posterior Probability Estimation A Theoretical Perspective 关于类后验概率估计焦点损失的理论视角 Semi-Supervised Synthesis of High-Resolution Editable Textures for 3D Humans 用于 3D 人体的高分辨率可以辑纹理的半监督合成 Transformer Interpretability Beyond Attention Visualization 超越注意力可视化的 Transformer 可解释性 How Privacy-Preserving Are Line Clouds Recovering Scene Details From 3D 线云如何从 3D 中恢复场景细节以保护隐私 Adaptive Image Transformer for One-Shot Object Detection 用于一次性目标检测的自适应图像转换器 AQD Towards Accurate Quantized Object Detection AQD 迈向准确的量化目标检测 Blind Deblurring for Saturated Images 饱和图像的盲去模糊 Camera-Space Hand Mesh Recovery via Semantic Aggregation and Adaptive 2D-1D 通过语义聚合和自适应 2D-1D 进行相机空间手网格恢复 Class-Aware Robust Adversarial Training for Object Detection 用于对象检测的类感知鲁棒对抗训练 Contrastive Neural Architecture Search With Neural Architecture Comparators 使用神经架构比较器进行对比神经架构搜索 DECOR-GAN 3D Shape Detailization by Conditional Refinement 条件细化的 DECOR-GAN 3D 形状细节化 Deep Analysis of CNN-Based Spatio-Temporal Representations for Action Recognition 深度分析基于 CNN 的时空表示以进行动作识别 Deep Texture Recognition via Exploiting Cross-Layer Statistical Self-Similarity 利用跨层统计自相似性进行深度纹理识别 Delving Deep Into Many-to-Many Attention for Few-Shot Video Object Segmentation 深入研究少镜头视频对象分割的多对多注意力 Distilling Audio-Visual Knowledge by Compositional Contrastive Learning 通过组合对比学习提炼视听知识 Distilling Knowledge via Knowledge Review 通过知识回顾提炼知识 DualAST Dual Style-Learning Networks for Artistic Style Transfer 用于艺术风格迁移的 DualAST 双风格学习网络 Dynamic Region-Aware Convolution 动态区域感知卷积 ECKPN Explicit Class Knowledge Propagation Network for Transductive Few-Shot Learning 用于转导小样本学习的 ECKPN 显式类知识传播网络 Efficient Object Embedding for Spliced Image Retrieval 用于拼接图像检索的高效对象嵌入 Equivariant Point Network for 3D Point Cloud Analysis 用于 3D 点云分析的等变点网络 Exploring Simple Siamese Representation Learning 探索简单的连体表示学习 FS-Net Fast Shape-Based Network for Category-Level 6D Object Pose Estimation 用于类别级 6D 对象姿态估计的 FS-Net 快速基于形状的网络 GeoSim Realistic Video Simulation via Geometry-Aware Composition for Self-Driving GeoSim 通过几何感知组合实现自动驾驶的真实视频模拟 High-Fidelity Face Tracking for ARVR via Deep Lighting Adaptation 通过深度照明适应实现 ARVR 的高保真人脸跟踪 Human-Like Controllable Image Captioning With Verb-Specific Semantic Roles 具有特定于动词的语义角色的类人可控图像字幕 Hybrid Rotation Averaging A Fast and Robust Rotation Averaging Approach 混合旋转平均 一种快速且稳健的旋转平均方法 I3Net Implicit Instance-Invariant Network for Adapting One-Stage Object Detectors I3Net 隐式实例不变网络适用于单阶段目标检测器 Indoor Lighting Estimation Using an Event Camera 使用事件相机进行室内照明估计 Jigsaw Clustering for Unsupervised Visual Representation Learning 用于无监督视觉表示学习的拼图聚类 Joint Generative and Contrastive Learning for Unsupervised Person Re-Identification 用于无监督人员重新识别的联合生成和对比学习 Learning 3D Shape Feature for Texture-Insensitive Person Re-Identification 为纹理不敏感的人重新识别学习 3D 形状特征 Learning a Non-Blind Deblurring Network for Night Blurry Images 学习用于夜间模糊图像的非盲去模糊网络 Learning Continuous Image Representation With Local Implicit Image Function 使用局部隐式图像函数学习连续图像表示 Learning Feature Aggregation for Deep 3D Morphable Models 深度 3D 可变形模型的学习特征聚合 Learning Student Networks in the Wild 在野外学习学生网络 Learning the Best Pooling Strategy for Visual Semantic Embedding 学习视觉语义嵌入的最佳池化策略 Localizing Visual Sounds the Hard Way 本地化视觉听起来很困难 MagDR Mask-Guided Detection and Reconstruction for Defending Deepfakes 用于防御 Deepfake 的 MagDR 掩模引导检测和重建 Model-Based 3D Hand Reconstruction via Self-Supervised Learning 通过自我监督学习进行基于模型的 3D 手部重建 MonoRUn Monocular 3D Object Detection by Reconstruction and Uncertainty Propagation MonoRUn 通过重构和不确定性传播的单目 3D 对象检测 Neural Feature Search for RGB-Infrared Person Re-Identification 用于 RGB 红外人重新识别的神经特征搜索 One-Shot Neural Ensemble Architecture Search by Diversity-Guided Search Space Shrinking 通过多样性引导搜索空间收缩的 One-Shot 神经集成架构搜索 Pareto Self-Supervised Training for Few-Shot Learning Perceptual Indistinguishability-Net PI-Net Facial Image Obfuscation With Manipulable Semantics 具有可操作语义的面部图像混淆 Points As Queries Weakly Semi-Supervised Object Detection by Points 点作为查询弱半监督对象检测点 Predicting Human Scanpaths in Visual Question Answering 在视觉问答中预测人类扫描路径 Pre-Trained Image Processing Transformer 预训练的图像处理转换器 Progressive Semantic-Aware Style Transformation for Blind Face Restoration 用于盲人脸恢复的渐进式语义感知风格转换 PSD Principled Synthetic-to-Real Dehazing Guided by Physical Priors 由物理先验引导的 PSD 原则合成到真实去雾 Reformulating HOI Detection As Adaptive Set Prediction 将 HOI 检测重新定义为自适应集预测 Robust and Accurate Object Detection via Adversarial Learning 通过对抗性学习进行稳健且准确的目标检测 Robust Representation Learning With Feedback for Single Image Deraining 具有反馈的鲁棒表示学习，用于单幅图像去雨 S2R-DepthNet Learning a Generalizable Depth-Specific Structural Representation S2R-DepthNet 学习可泛化的深度特定结构表示 Scale-Aware Automatic Augmentation for Object Detection 用于对象检测的规模感知自动增强 Scan2Cap Context-Aware Dense Captioning in RGB-D Scans RGB-D 扫描中的 Scan2Cap 上下文感知密集字幕 Scene Text Telescope Text-Focused Scene Image Super-Resolution 场景文本望远镜文本聚焦场景图像超分辨率 Semantic Audio-Visual Navigation 语义视听导航 Semi-Supervised Domain Adaptation Based on Dual-Level Domain Mixing for Semantic 基于双级域混合的语义半监督域自适应 Semi-Supervised Semantic Segmentation With Cross Pseudo Supervision 具有交叉伪监督的半监督语义分割 Shot Contrastive Self-Supervised Learning for Scene Boundary Detection 用于场景边界检测的镜头对比自监督学习 The Lottery Tickets Hypothesis for Supervised and Self-Supervised Pre-Training in 监督和自我监督预训练的彩票假说 Topological Planning With Transformers for Vision-and-Language Navigation 用于视觉和语言导航的变压器拓扑规划 Towards Bridging Event Captioner and Sentence Localizer for Weakly Supervised Dense Event Captioning面向弱监督密集事件字幕的桥接事件字幕和句子定位器 Transformer Tracking 变压器跟踪 Triple-Cooperative Video Shadow Detection 三重协作视频阴影检测 Wasserstein Contrastive Representation Distillation Wasserstein 对比表示蒸馏法 Wide-Baseline Relative Camera Pose Estimation With Directional Learning 具有方向学习的宽基线相对相机姿态估计 You Only Look One-Level Feature 你只看一级特征 (AF)2-S3Net: Attentive Feature Fusion with Adaptive Feature Selection for Sparse Semantic Segmentation Network稀疏语义分割网络中具有自适应特征选择的注意力特征融合 Back-Tracing Representative Points for Voting-Based 3D Object Detection in Point 点中基于投票的 3D 对象检测的回溯代表点 Boundary IoU Improving Object-Centric Image Segmentation Evaluation 边界 IoU 改进以对象为中心的图像分割评估 Learning Deep Classifiers Consistent With Fine-Grained Novelty Detection 学习与细粒度新奇检测一致的深度分类器 Learning To Filter Siamese Relation Network for Robust Tracking 学习过滤孪生关系网络以实现鲁棒跟踪 Light Field Super-Resolution With Zero-Shot Learning 具有零样本学习的光场超分辨率 Memory-Efficient Network for Large-Scale Video Compressive Sensing 用于大规模视频压缩感知的内存高效网络 Modular Interactive Video Object Segmentation Interaction-to-Mask Propagation and Difference-Aware Fusion 模块化交互式视频对象分割 Interaction-to-Mask 传播和差异感知融合 Monocular 3D Multi-Person Pose Estimation by Integrating Top-Down and Bottom-Up 通过集成自顶向下和自底向上的单目 3D 多人姿势估计 Multi-View 3D Reconstruction of a Texture-Less Smooth Surface of Unknown 未知的无纹理光滑表面的多视图 3D 重建 NBNet Noise Basis Learning for Image Denoising With Subspace Projection 基于子空间投影的图像去噪的 NBNet 噪声基础学习 Style-Aware Normalized Loss for Improving Arbitrary Style Transfer 用于改进任意风格迁移的风格感知归一化损失 Semantic-Aware Knowledge Distillation for Few-Shot Class-Incremental Learning 少样本增量学习的语义感知知识蒸馏 Navigating the GAN Parameter Space for Semantic Image Editing 为语义图像编辑导航 GAN 参数空间 Feature-Level Collaboration: Joint Unsupervised Learning of Optical Flow,Stereo Depth and Camera Motion特征级协作：光流、立体深度和相机运动的联合无监督学习 Test-Time Fast Adaptation for Dynamic Scene Deblurring via Meta-Auxiliary Learning 基于元辅助学习的动态场景去模糊测试时间快速适应 Stereo Radiance Fields (SRF): Learning View Synthesis for Sparse Views of Novel Scenes新颖场景稀疏视图的学习型视图合成 PiCIE Unsupervised Semantic Segmentation Using Invariance and Equivariance in Clustering 在聚类中使用不变性和等变性进行 PiCIE 无监督语义分割 Beyond Static Features for Temporally Consistent 3D Human Pose and Shape from a Video超越静态特征，用于视频时间一致的3D人类姿势和形状估计 Meta Batch-Instance Normalization for Generalizable Person Re-Identification RobustNet Improving Domain Generalization in Urban-Scene Segmentation via Instance Selective RobustNet 通过实例选择改进城市场景分割中的域泛化 Shared Cross-Modal Trajectory Prediction for Autonomous Driving 自动驾驶的共享跨模态轨迹预测 VaB-AL Incorporating Class Imbalance and Difficulty With Variational Bayes for VaB-AL 将类不平衡和难度与变分贝叶斯相结合 VITON-HD High-Resolution Virtual Try-On via Misalignment-Aware Normalization VITON-HD 高分辨率虚拟试穿通过错位感知归一化 Mask-ToF Learning Microlens Masks for Flying Pixel Correction in Time-of-Flight Imaging学习微透镜掩模以在飞行时间成像中进行飞行像素校正 Probabilistic Embeddings for Cross-Modal Retrieval 跨模态检索的概率嵌入 Multi-Label Learning From Single Positive Labels 从单个正标签进行多标签学习 Correlated Input-Dependent Label Noise in Large-Scale Image Classification 大规模图像分类中的相关输入相关标签噪声 Differentiable Patch Selection for Image Recognition 图像识别的可微分块选择 SMPLicit Topology-Aware Generative Model for Clothed People 有衣人的拓扑感知生成模型 Zillow Indoor Dataset Annotated Floor Plans With 360deg Panoramas and 3D Room LayoutsZillow 室内数据集 带有 360 度全景图和 3D 房间布局的带注释的平面图 Asymmetric Gained Deep Image Compression With Continuous Rate Adaptation具有连续速率自适应的非对称获得深度图像压缩 GANmut Learning Interpretable Conditional Space for Gamut of Emotions 学习情绪范围的可解释条件空间 Geus Part-Aware Panoptic SegmentationGeus 部分感知全景分割 Animating Pictures With Eulerian Motion Fields 用欧拉运动场动画图片 Composing Photos Like a Photographer 像摄影师一样构图 Disentangling Label Distribution for Long-Tailed Visual Recognition 解开长尾视觉识别的标签分布 Fine-Grained Shape-Appearance Mutual Learning for Cloth-Changing Person Re-Identification 细粒度形貌互学习换衣人再识别 LiDAR-Based Panoptic Segmentation via Dynamic Shifting Network 基于 LiDAR 的动态移动网络全景分割 LPSNet A Lightweight Solution for Fast Panoptic Segmentation LPSNet 一种用于快速全景分割的轻量级解决方案 Panoramic Image Reflection Removal全景图像反射去除 Reinforced Attention for Few-Shot Learning and Beyond 加强对 Few-Shot 学习及其他学习的关注 StereoPIFu Depth Aware Clothed Human Digitization via Stereo Vision StereoPIFu 深度感知穿衣人体数字化通过立体视觉 Student-Teacher Learning From Clean Inputs to Noisy Inputs 师生从干净输入到嘈杂输入的学习 StyleMix Separating Content and Style for Enhanced Data Augmentation StyleMix 分离内容和样式以增强数据增强 Transformation Driven Visual Reasoning 转换驱动的视觉推理 VLN BERT A Recurrent Vision-and-Language BERT for Navigation VLN BERT 用于导航的循环视觉和语言 BERT DSRNA Differentiable Search of Robust Neural Architectures 稳健神经架构的 DSRNA 可微搜索 Image Change Captioning by Learning From an Auxiliary Task 通过辅助任务学习图像更改字幕 Affordance Transfer Learning for Human-Object Interaction Detection 人与物体交互检测的可供性迁移学习 BiCnet-TKS Learning Efficient Spatial-Temporal Representation for Video Person Re-Identification BiCnet-TKS 学习用于视频人物重新识别的高效时空表示 Coordinate Attention for Efficient Mobile Network Design 协调注意力以实现高效的移动网络设计 Detecting Human-Object Interaction via Fabricated Compositional Learning 通过虚构的组合学习检测人与物体的交互 Exploring Data-Efficient 3D Scene Understanding With Contrastive Scene Contexts 使用对比场景上下文探索数据高效的 3D 场景理解 Informative and Consistent Correspondence Mining for Cross-Domain Weakly Supervised Object 跨域弱监督对象的信息一致对应挖掘 Towards High Fidelity Face Relighting With Realistic Shadows 使用逼真的阴影实现高保真面部重新照明 Visualizing Adapted Knowledge in Domain Transfer 可视化领域迁移中的适应知识 Three Ways To Improve Semantic Segmentation With Self-Supervised Depth Estimation 使用自监督深度估计改进语义分割的三种方法 DARCNN Domain Adaptive Region-Based Convolutional Neural Network for Unsupervised Instance 用于无监督实例的 DARCNN 域自适应区域卷积神经网络 A2-FPN Attention Aggregation Based Feature Pyramid Network for Instance Segmentation 用于实例分割的基于 A2-FPN 注意力聚合的特征金字塔网络 AdCo: Adversarial Contrast for Efficient Learning of Unsupervised Representations from Self-Trained Negative AdversariesAdCo：有效学习自训练的负面对手的无监督表示的对抗对比 Bidirectional Projection Network for Cross Dimension Scene Understanding 用于跨维度场景理解的双向投影网络 Dense Relation Distillation With Context-Aware Aggregation for Few-Shot Object Detection 具有上下文感知聚合的密集关系蒸馏，用于少镜头目标检测 Distilling Causal Effect of Data in Class-Incremental Learning 在类增量学习中提取数据的因果效应 Efficient Deformable Shape Correspondence via Multiscale Spectral Manifold Wavelets Preservation 基于多尺度谱流形小波保存的高效可变形形状对应 FVC A New Framework Towards Deep Video Compression in Feature SpaceFVC 一个面向特征深度视频压缩的新框架 Learning Cross-Modal Retrieval With Noisy Labels 学习带噪声标签的跨模态检索 Learning Position and Target Consistency for Memory-Based Video Object Segmentation 基于内存的视频对象分割的学习位置和目标一致性 Model-Aware Gesture-to-Gesture Translation 模型感知手势到手势转换 Pseudo 3D Auto-Correlation Network for Real Image Denoising 用于真实图像去噪的伪 3D 自相关网络 Safe Local Motion Planning With Self-Supervised Freespace Forecasting 具有自我监督自由空间预测的安全局部运动规划 SAIL-VOS 3D A Synthetic Dataset and Baselines for Object Detection SAIL-VOS 3D 目标检测的合成数据集和基线 Self-Supervised 3D Mesh Reconstruction From Single Images 从单幅图像进行自我监督 3D 网格重建 SimPLE Similar Pseudo Label Exploitation for Semi-Supervised Classification 用于半监督分类的 SimPLE 相似伪标签开发 Towards Semantic Segmentation of Urban-Scale 3D Point Clouds A Dataset 迈向城市规模 3D 点云的语义分割数据集 Wide-Depth-Range 6D Object Pose Estimation in Space 空间中的宽深度范围 6D 对象姿态估计 A Multiplexed Network for End-to-End Multilingual OCR 用于端到端多语言 OCR 的多路复用网络 Brain Image Synthesis With Unsupervised Multivariate Canonical CSCl4Net 使用无监督多元规范 CSCl4Net 进行脑图像合成 Cross-View Regularization for Domain Adaptive Panoptic Segmentation 域自适应全景分割的跨视图正则化 Deep Gaussian Scale Mixture Prior for Spectral Compressive Imaging 用于光谱压缩成像的深高斯尺度混合先验 DeepLM Large-Scale Nonlinear Least Squares on Deep Learning Frameworks Using DeepLM 在深度学习框架上使用的大规模非线性最小二乘法 DI-Fusion Online Implicit 3D Reconstruction With Deep Priors 具有深度先验的 DI-Fusion 在线隐式 3D 重建 Few-Shot Human Motion Transfer by Personalized Geometry and Texture Modeling 个性化几何和纹理建模的小样本人体运动传输 FSDR Frequency Space Domain Randomization for Domain Generalization 用于域泛化的 FSDR 频域域随机化 Geo-FARM Geodesic Factor Regression Model for Misaligned Pre-Shape Responses in Geo-FARM 测地线因子回归模型，用于未对齐的预形状响应 Group Whitening Balancing Learning Efficiency and Representational Capacity 群体美白：平衡学习效率和表征能力 Look Before You Leap Learning Landmark Features for One-Stage Visual Grounding跳之前先看看：学习用于单阶段视觉基础的地标特征 Memory Oriented Transfer Learning for Semi-Supervised Image Deraining 半监督图像去雨的面向记忆的迁移学习 MetaSets Meta-Learning on Point Sets for Generalizable Representations MetaSets 基于点集的元学习用于泛化表示 MetricOpt Learning To Optimize Black-Box Evaluation Metrics MetricOpt 学习优化黑盒评估指标 MOS Towards Scaling Out-of-Distribution Detection for Large Semantic Space MOS 面向扩展大语义空间的分布外检测 MultiBodySync Multi-Body Segmentation and Motion Estimation via 3D Scan Synchronization MultiBodySync 通过 3D 扫描同步进行多体分割和运动估计 Neighbor2Neighbor Self-Supervised Denoising From Single Noisy Images Neighbor2Neighbor 从单个噪声图像中进行自我监督去噪 Predator Registration of 3D Point Clouds With Low Overlap 低重叠 3D 点云的捕食者配准 Revisiting Knowledge Distillation An Inheritance and Exploration Framework 重温知识蒸馏的继承和探索框架 S3 Learnable Sparse Signal Superdensity for Guided Depth Estimation 用于引导深度估计的 S3 可学习稀疏信号超密度 Searching by Generating Flexible and Efficient One-Shot NAS With Architecture 通过生成具有架构的灵活高效的 One-Shot NAS 进行搜索 Seeing Out of the Box End-to-End Pre-Training for Vision-Language Representation 开箱即用的视觉语言表示端到端预训练 Self-Supervised Motion Learning From Static Images 从静态图像中进行自我监督运动学习 Self-Supervised Video Representation Learning by Context and Motion Decoupling 基于上下文和运动解耦的自监督视频表示学习 Video Rescaling Networks With Joint Optimization Strategies for Downscaling and Upscaling具有用于缩小和放大的联合优化策略的视频重新缩放网络 VS-Net Voting With Segmentation for Visual Localization视觉定位的分割投票 When Age-Invariant Face Recognition Meets Face Age Synthesis： A Multi-Task Learning Framework当年龄不变的人脸识别遇到人脸年龄合成：一项多任务学习框架 Collaborative Spatial-Temporal Modeling for Language-Queried Video Actor Segmentation 用于语言查询视频演员分割的协作时空建模 Learning the Non-Differentiable Optimization for Blind Super-Resolution 学习盲超分辨率的不可微优化 ATSO Asynchronous Teacher-Student Optimization for Semi-Supervised Image Segmentation ATSO 异步师生优化半监督图像分割 Self-Supervised Multi-Frame Monocular Scene Flow 自监督多帧单目场景流 Progressive Semantic Segmentation 渐进式语义分割 Exemplar-Based Open-Set Panoptic Segmentation Network 基于示例的开放集全景分割网络 Self-Supervised Video GANs Learning for Appearance Consistency and Motion Coherency 自我监督视频 GAN 学习外观一致性和运动连贯性 3D Shape Generation With Grid-Based Implicit Functions 使用基于网格的隐式函数生成 3D 形状 Shape From Sky Polarimetric Normal Recovery Under the Sky 天空下的极化法线恢复形状 Optimal Quantization Using Scaled Codebook 使用缩放码本的最佳量化 Depth Completion With Twin Surface Extrapolation at Occlusion Boundaries 在遮挡边界处使用双曲面外推的深度补全 Passive Inter-Photon Imaging 被动光子间成像 Multi-Target Domain Adaptation With Collaborative Consistency Learning 具有协作一致性学习的多目标域适应 Facial Action Unit Detection With Transformers 使用变形金刚进行面部动作单元检测 Learning High Fidelity Depths of Dressed Humans by Watching Social 通过观看社交来学习穿着打扮的人的高保真深度 KeypointDeformer Unsupervised 3D Keypoint Discovery for Shape Control 用于形状控制的无监督 3D 关键点发现 CAMERAS Enhanced Resolution and Sanity Preserving Class Activation Mapping for image saliency图像显着性的增强分辨率和健全性保留类激活映射 MeanShift Extremely Fast Mode-Seeking With Applications to Segmentation and Object Tracking用于分割和对象的 MeanShift 极快模式搜索 NewtonianVAE Proportional Control and Goal Identification From Pixels via Physical NewtonianVAE 比例控制和通过物理从像素识别目标 Quantifying Explainers of Graph Neural Networks in Computational Pathology 量化计算病理学中图神经网络的解释器 UV-Net Learning From Boundary Representations UV-Net 从边界表示中学习 Mining Better Samples for Contrastive Learning of Temporal Correspondence 挖掘更好的样本用于时间对应的对比学习 Few-Shot Open-Set Recognition by Transformation Consistency 基于变换一致性的 Few-Shot 开集识别 Interpolation-Based Semi-Supervised Learning for Object Detection 用于目标检测的基于插值的半监督学习 Memory-Guided Unsupervised Image-to-Image Translation记忆引导的无监督图像到图像转换 Audio-Driven Emotional Video Portraits 音频驱动的情感视频肖像 Calibrated RGB-D Salient Object Detection 校准的 RGB-D 显着目标检测 Learning Calibrated Medical Image Segmentation via Multi-Rater Agreement Modeling 通过多评价者协议建模学习校准的医学图像分割 Refine Myself by Teaching Myself Feature Refinement via Self-Knowledge Distillation 通过自我知识蒸馏自学特征细化来细化自己 Intentonomy A Dataset and Study Towards Human Intent Understanding Intentonomy 一个数据集和对人类意图理解的研究 IoU Attack Towards Temporally Coherent Black-Box Adversarial Attack for Visual IoU 攻击对视觉的时间相干黑盒对抗攻击 Leveraging Line-Point Consistence To Preserve Structures for Wide Parallax Image 利用线点一致性保留宽视差图像的结构 Scalability vs. Utility Do We Have To Sacrifice One for 可扩展性与实用性我们是否必须为此牺牲一个？ Learning Compositional Representation for 4D Captures With Neural ODE 使用神经 ODE 学习 4D 捕获的组合表示 Learning Optical Flow From a Few Matches 从几场比赛中学习光流 Regressive Domain Adaptation for Unsupervised Keypoint Detection 无监督关键点检测的回归域自适应 Robust Reference-Based Super-Resolution via C2-Matching 通过 C2 匹配实现强大的基于参考的超分辨率 Saliency-Guided Image Translation 显着性引导的图像翻译 EffiScene Efficient Per-Pixel Rigidity Inference for Unsupervised Joint Learning of Optical Flow, Depth, Camera Pose and Motion Segmentation用于光流、深度、相机姿势和运动分割的无监督联合学习的高效逐像素刚性推断 Harmonious Semantic Line Detection via Maximal Weight Clique Selection 基于最大权重筛选的调和语义分割检测 Teachers Do More Than Teach Compressing Image-to-Image Models 教师所做的不仅仅是教授压缩图像到图像模型 Amalgamating Knowledge From Heterogeneous Graph Neural Networks 融合来自异构图神经网络的知识 Cross-Modal Center Loss for 3D Cross-Modal Retrieval 用于 3D 跨模态检索的跨模态中心损失 Locate Then Segment A Strong Pipeline for Referring Image Segmentation 定位然后分割一个强大的管道用于参考图像分割 Turning Frequency to Resolution Video Super-Resolution via Event Cameras 通过事件摄像机将频率转换为分辨率视频超分辨率 Practical Single-Image Super-Resolution Using Look-Up Table 使用查找表的实用单图像超分辨率 Tackling the Ill-Posedness of Super-Resolution Through Adaptive Target Generation 通过自适应目标生成解决超分辨率的弊端 Towards Open World Object Detection 迈向开放世界目标检测 Joint Deep Model-Based MR Image and Coil Sensitivity Reconstruction Network 基于联合深度模型的 MR 图像和线圈灵敏度重建网络 Fair Feature Distillation for Visual Recognition 视觉识别的公平特征蒸馏 Time Adaptive Recurrent Neural Network 时间自适应递归神经网络 Coarse-Fine Networks for Temporal Activity Detection in Videos 用于视频中时间活动检测的粗细网络 In the Light of Feature Distributions Moment Matching for Neural 根据神经网络的特征分布矩匹配 Relative Order Analysis and Optimization for Unsupervised Deep Metric Learning 无监督深度度量学习的相对阶分析和优化 Blur Noise and Compression Robust Generative Adversarial Networks 模糊噪声和压缩鲁棒生成对抗网络 Unsupervised Learning of Depth and Depth-of-Field Effect From Natural Images with Aperture Rendering Generative Adversarial Networks使用孔径渲染生成对抗网络从自然图像中进行深度和景深效应的无监督学习 Guided Integrated Gradients An Adaptive Path Method for Removing Noise 引导积分梯度一种自适应路径去除噪声的方法 High-Fidelity Neural Human Motion Transfer From Monocular Video 单目视频的高保真神经人体运动传输 Fast Bayesian Uncertainty Estimation and Reduction of Batch Normalized Single 批量归一化单的快速贝叶斯不确定性估计和减少 Zero-Shot Single Image Restoration Through Controlled Perturbation of Koschmieders Model 通过 Koschmieders 模型的受控扰动实现零样本单图像恢复 MAZE Data-Free Model Stealing Attack Using Zeroth-Order Gradient Estimation 使用零阶梯度估计的 MAZE 无数据模型窃取攻击 Differentiable SLAM-Net Learning Particle SLAM for Visual Navigation 用于视觉导航的微分 SLAM-Net 学习粒子 SLAM Uncalibrated Neural Inverse Rendering for Photometric Stereo of General Surfaces 一般表面光度立体的未校准神经逆向渲染 Deep Occlusion-Aware Instance Segmentation With Overlapping BiLayers 具有重叠双层的深度遮挡感知实例分割 Neural Lumigraph Rendering 神经 Lumigraph 渲染 Hierarchical Lovasz Embeddings for Proposal-Free Panoptic Segmentation 用于无提议全景分割的分层 Lovasz 嵌入 How Transferable Are Reasoning Patterns in VQA VQA 中的推理模式如何转移 Roses Are Red Violets Are Blue… but Should VQA Expect 玫瑰是红紫罗兰是蓝色的……但 VQA 应该期待吗 Neural Response Interpretation Through the Lens of Critical Pathways 从关键路径的角度解读神经反应 Differentiable Diffusion for Dense Depth Estimation From Multi-View Images 多视图图像密集深度估计的微分扩散 UniT Unified Knowledge Transfer for Any-Shot Object Detection and Segmentation 用于 Any-Shot 对象检测和分割的 UnitT 统一知识转移 Neural Side-by-Side Predicting Human Preferences for No-Reference Super-Resolution Evaluation 用于无参考超分辨率评估的神经并排预测人类偏好 Discriminative Appearance Modeling With Multi-Track Pooling for Real-Time Multi-Object Tracking 用于实时多目标跟踪的多轨道池的判别外观建模 DriveGAN Towards a Controllable High-Quality Neural Simulation DriveGAN 迈向可控的高质量神经仿真 Embedding Transfer With Label Relaxation for Improved Metric Learning 嵌入带有标签松弛的迁移以改进度量学习 Exploiting Spatial Dimensions of Latent in GAN for Real-Time Image 在实时图像中利用 GAN 中潜在的空间维度 High-Quality Stereo Image Restoration From Double Refraction 双折射的高质量立体图像恢复 HOTR End-to-End Human-Object Interaction Detection With Transformers HOTR 使用 Transformer 的端到端人与物体交互检测 Improving Accuracy of Binary Neural Networks Using Unbalanced Activation Distribution 使用不平衡激活分布提高二元神经网络的准确性 IronMask Modular Architecture for Protecting Deep Face Template IronMask 模块化架构保护深面模板 Joint Negative and Positive Learning for Noisy Labels 噪声标签的联合消极和积极学习 KOALAnet Blind Super-Resolution Using Kernel-Oriented Adaptive Local Adjustment 使用面向内核的自适应局部调整的 KOALAnet 盲超分辨率 LaPred Lane-Aware Prediction of Multi-Modal Future Trajectories of Dynamic Agents 动态智能体多模态未来轨迹的车道感知预测 Not Just Compete but Collaborate Local Image-to-Image Translation via Cooperative 不仅仅是竞争，而是通过合作进行本地图像到图像的翻译 Prototype-Guided Saliency Feature Learning for Person Search 用于人员搜索的原型引导显着性特征学习 Quality-Agnostic Image Recognition via Invertible Decoder 通过可逆解码器的质量不可知图像识别 SetVAE Learning Hierarchical Composition for Generative Modeling of Set-Structured Data 用于集合结构数据生成建模的 SetVAE 学习分层组合 Task-Aware Variational Adversarial Active Learning 任务感知变分对抗主动学习 XProtoNet Diagnosis in Chest Radiography With Global and Local Explanations XProtoNet 胸部 XProtoNet 诊断与全局和局部解释 FlowStep3D Model Unrolling for Self-Supervised Scene Flow Estimation 用于自监督场景流估计的 FlowStep3D 模型展开 How To Exploit the Transferability of Learned Image Compression to 如何利用学习图像压缩的可迁移性 Cuboids Revisited Learning Robust 3D Shape Fitting to Single RGB ImagesCuboids 重新审视学习稳健的 3D 形状拟合到单个 RGB图像 T-vMF Similarity for Regularizing Intra-Class Feature Distribution 用于正则化类内特征分布的 T-vMF 相似性 Learning Monocular 3D Reconstruction of Articulated Categories From Motion 从运动中学习铰接类别的单目 3D 重建 MoViNets Mobile Video Networks for Efficient Video Recognition 用于高效视频识别的 MoViNets 移动视频网络 ClassSR A General Framework to Accelerate Super-Resolution Networks by Data ClassSR 一个通过数据加速超分辨率网络的通用框架 Robust Consistent Video Depth Estimation 稳健一致的视频深度估计 Interpretable Social Anchors for Human Trajectory Forecasting in Crowds 人群中人类轨迹预测的可解释社会锚 Weakly-Supervised Physically Unconstrained Gaze Estimation 弱监督物理无约束注视估计 Rethinking Style Transfer From Pixels to Parameterized Brushstrokes 重新思考从像素到参数化笔触的风格转换 QPP Real-Time Quantization Parameter Prediction for Deep Neural Networks 深度神经网络的 QPP 实时量化参数预测 Hierarchical Motion Understanding via Motion Programs 通过运动程序理解分层运动 GrooMeD-NMS Grouped Mathematically Differentiable NMS for Monocular 3D Object Detection GrooMeD-NMS 用于单目 3D 目标检测的分组数学可微分 NMS Controllable Image Restoration for Under-Display Camera in Smartphones 智能手机屏下摄像头的可控图像恢复 Single-View Robot Pose and Joint Angle Estimation via Render 通过渲染进行单视图机器人姿态和关节角度估计 IMODAL Creating Learnable User-Defined Deformation Models IMODAL 创建可学习的用户定义变形模型 LipSync3D Data-Efficient Learning of Personalized 3D Talking Faces From Video LipSync3D 数据高效学习视频中的个性化 3D 会说话的面孔 Semi-Supervised Semantic Segmentation With Directional Context-Aware Consistency 具有方向上下文感知一致性的半监督语义分割 CoCoNets Continuous Contrastive 3D Scene Representations CoCoNets 连续对比 3D 场景表示 Restoring Extremely Dark Images in Real Time 实时恢复极暗图像 BRepNet A Topological Message Passing System for Solid Models BRepNet 实体模型的拓扑消息传递系统 General Multi-Label Image Classification With Transformers 使用 Transformer 的通用多标签图像分类 Pulsar Efficient Sphere-Based Neural Rendering Pulsar 高效的基于球体的神经渲染 Moing Semantic Palette Guiding Scene Generation With Class Proportions Moing 语义调色板用类比例指导场景生成 MongeNet Efficient Sampler for Geometric Deep Learning 用于几何深度学习的 MongeNet 高效采样器 3D Video Stabilization With Depth Estimation by CNN-Based Optimization 通过基于 CNN 的优化进行深度估计的 3D 视频稳定 Anti-Adversarially Manipulated Attributions for Weakly and Semi-Supervised Semantic Segmentation 弱和半监督语义分割的反对抗操纵属性 BBAM Bounding Box Attribution Map for Weakly Supervised Semantic and Instance Segmentation用于弱监督语义和实例分割的 BBAM 边界框属性图 Blocks-World Cameras Blocks-世界相机 CoSMo Content-Style Modulation for Image Retrieval With Text Feedback 带有文本反馈的图像检索的 CoSMo 内容样式调制 Depth Completion Using Plane-Residual Representation 使用平面残差表示完成深度 DRANet Disentangling Representation and Adaptation Networks for Unsupervised Cross-Domain Adaptation 用于无监督跨域适应的 DRANet 解开表示和适应网络 Iterative Filter Adaptive Network for Single Image Defocus Deblurring 用于单图像散焦去模糊的迭代滤波器自适应网络 Large-Scale Localization Datasets in Crowded Indoor Spaces 拥挤室内空间中的大规模定位数据集 Looking Into Your Speech Learning Cross-Modal Affinity for Audio-Visual Speech Separation调查您的语音学习视听语音的跨模式亲和力 Network Quantization With Element-Wise Gradient Scaling 使用逐元素梯度缩放的网络量化 PatchMatch-Based Neighborhood Consensus for Semantic Correspondence 基于 PatchMatch 的语义对应邻域共识 Railroad Is Not a Train Saliency As Pseudo-Pixel Supervision for 铁路不是作为伪像素监督的火车显着性 Regularization Strategy for Point Cloud via Rigidly Mixed Sample 基于刚性混合样本的点云正则化策略 Relevance-CAM Your Model Already Knows Where To Look Relevance-CAM 你的模型已经知道去哪里找 Restore From Restored Video Restoration With Pseudo Clean Video 使用伪干净视频从恢复的视频恢复中恢复 Rotation-Only Bundle Adjustment 仅旋转捆绑调整 SIPSA-Net： Shift-Invariant Pan Sharpening With Moving Object Alignment for Satellite 与卫星移动对象对齐 Video Prediction Recalling Long-Term Motion Context via Memory Alignment Learning 通过记忆对齐学习回忆长期运动上下文的视频预测 Less Is More ClipBERT for Video-and-Language Learning via Sparse Sampling 少即是多 ClipBERT 通过稀疏采样进行视频和语言学习 Picasso A CUDA-Based Library for Deep Learning Over 3D Meshes Picasso 基于 CUDA 的 3D 网格深度学习库 Robust Reflection Removal With Reflection-Free Flash-Only Cues 使用无反射仅闪光提示的强大反射去除 2D or not 2D Adaptive 3D Convolution Selection for Efficient Video Recognition 用于高效视频识别的 2D 或非 2D 自适应 3D 卷积选择 3D Human Action Representation Learning via Cross-View Consistency Pursuit 通过跨视图一致性追求的 3D 人类行为表示学习 Action Shuffle Alternating Learning for Unsupervised Action Segmentation 无监督动作分割的动作洗牌交替学习 Adaptive Prototype Learning and Allocation for Few-Shot Segmentation 少镜头分割的自适应原型学习和分配 Anchor-Constrained Viterbi for Set-Supervised Action Segmentation 用于集合监督动作分割的锚约束维特比 ARVo Learning All-Range Volumetric Correspondence for Video Deblurring ARVo 学习用于视频去模糊的全范围体积对应 Beyond Max-Margin Class Margin Equilibrium for Few-Shot Object Detection Bipartite Graph Network With Adaptive Message Passing for Unbiased Scene 具有自适应消息传递的无偏场景二分图网络 Causal Hidden Markov Model for Time Series Disease Forecasting 时间序列疾病预测的因果隐马尔可夫模型 Combined Depth Space Based Architecture Search for Person Re-Identification 用于人员重新识别的基于组合深度空间的架构搜索 Continuous Face Aging via Self-Estimated Residual Age Embedding 通过自估计剩余年龄嵌入进行连续人脸老化 Cross-Domain Adaptive Clustering for Semi-Supervised Domain Adaptation 用于半监督域自适应的跨域自适应聚类 CutPaste Self-Supervised Learning for Anomaly Detection and Localization 用于异常检测和定位的 CutPaste 自监督学习 D2IM-Net Learning Detail Disentangled Implicit Fields From Single Images D2IM-Net 学习细节从单个图像中分离出隐含字段 DeepI2P Image-to-Point Cloud Registration via Deep Classification 基于深度分类的 DeepI2P 图像到点云配准 Diverse Part Discovery Occluded Person Re-Identification With Part-Aware Transformer 使用 Part-Aware Transformer 对不同部分发现遮挡人员进行重新识别 Domain Consensus Clustering for Universal Domain Adaptation 通用域自适应的域共识聚类 Dual-Stream Multiple Instance Learning Network for Whole Slide Image Classification 用于全幻灯片图像分类的双流多实例学习网络 Dynamic Class Queue for Large Scale Face Recognition in the 用于大规模人脸识别的动态类队列 Dynamic Domain Adaptation for Efficient Inference 高效推理的动态域适应 Dynamic Slimmable Network 动态可精简网络 Dynamic Transfer for Multi-Source Domain Adaptation 多源域自适应的动态迁移 Ego-Exo Transferring Visual Representations From Third-Person to First-Person Videos Ego-Exo 将视觉表征从第三人称视频转移到第一人称视频 Exploring Adversarial Fake Images on Face Manifold 探索人脸歧管上的对抗性假图像 Exploring intermediate representation for monocular vehicle pose estimation 探索用于单目车辆姿态估计的中间表示 FaceInpainter High Fidelity Face Adaptation to Heterogeneous Domains FaceInpainter 高保真人脸适应异构域 Few-Shot Object Detection via Classification Refinement and Distractor Retreatment 通过分类细化和 Distractor Retreatment 进行 Few-Shot 目标检测 Frequency-Aware Discriminative Feature Learning Supervised by Single-Center Loss for Face 人脸单中心损失监督的频率感知判别特征学习 From Synthetic to Real Unsupervised Domain Adaptation for Animal Pose 从合成到真正的无监督领域适应动物姿势 Fully Convolutional Networks for Panoptic Segmentation 用于全景分割的全卷积网络 Generalized Focal Loss V2 Learning Reliable Localization Quality Estimation for Dense Object Detection学习用于密集对象检测的可靠定位质量估计 Generalizing to the Open World Deep Visual Odometry With Online Adaptation通过在线适应推广到开放世界深度视觉里程计 HCRF-Flow： Scene Flow From Point Clouds With Continuous High-Order CRFs 来自具有连续高阶 CRF 的点云的场景流 Hilbert Sinkhorn Divergence for Optimal Transport HybrIK A Hybrid Analytical-Neural Inverse Kinematics Solution for 3D Human Pose and Shape Estimation用于 3D 人体姿势和形状估计的混合分析神经逆运动学解决方案 Image-to-Image Translation via Hierarchical Style Disentanglement 通过分层样式解缠结实现图像到图像的转换 Involution Inverting the Inherence of Convolution for Visual Recognition 对卷积反转视觉识别的内在性 Learning Invariant Representations and Risks for Semi-Supervised Domain Adaptation 半监督域适应的学习不变表示和风险 Learning Probabilistic Ordinal Embeddings for Uncertainty-Aware Regression 学习用于不确定性感知回归的概率序数嵌入 Learning To Identify Correct 2D-2D Line Correspondences on Sphere 学习识别球体上正确的 2D-2D 线对应关系 LiDAR R-CNN An Efficient and Universal 3D Object Detector LiDAR R-CNN 一种高效且通用的 3D 物体检测器 Lighting Reflectance and Geometry Estimation From 360deg Panoramic Stereo 360度全景立体照明反射率和几何估计 Meta-Mining Discriminative Samples for Kinship Verification 用于亲属关系验证的元挖掘判别样本 MetaSAug Meta Semantic Augmentation for Long-Tailed Visual Recognition MetaSAug 用于长尾视觉识别的元语义增强 Model-Contrastive Federated Learning 模型对比联邦学习 Neural Scene Flow Fields for Space-Time View Synthesis of Dynamic 用于动态时空视图合成的神经场景流场 NPAS A Compiler-Aware Framework of Unified Network Pruning and Architecture NPAS 一个编译器感知的统一网络修剪和架构框架 On Feature Normalization and Data Augmentation 关于特征归一化和数据增强 OpenRooms An Open Framework for Photorealistic Indoor Scene Datasets OpenRooms 一个用于逼真的室内场景数据集的开放框架 Point Cloud Upsampling via Disentangled Refinement 通过分离细化进行点云上采样 PointFlow Flowing Semantics Through Points for Aerial Image Segmentation 用于航空图像分割的 PointFlow 流语义通过点 PointNetLK Revisited 重访PointNetLK Pose Recognition With Cascade Transformers 使用级联变压器进行姿势识别 POSEFusion Pose-Guided Selective Fusion for Single-View Human Volumetric Capture POSEFusion 用于单视图人体体积捕获的姿势引导选择性融合 Probabilistic Model Distillation for Semantic Correspondence 语义对应的概率模型蒸馏 Progressive Domain Expansion Network for Single Domain Generalization 单域泛化的渐进域扩展网络 Progressive Stage-Wise Learning for Unsupervised Feature Representation Enhancement 无监督特征表示增强的渐进式阶段学习 QAIR Practical Query-Efficient Black-Box Attacks for Image Retrieval 用于图像检索的 QAIR 实用高效查询黑盒攻击 Ranking Neural Checkpoints 对神经检查点进行排名 Representing Videos As Discriminative Sub-Graphs for Action Recognition 将视频表示为动作识别的判别子图 Searching for Fast Model Families on Datacenter Accelerators 在数据中心加速器上搜索快速模型系列 SelfDoc Self-Supervised Document Representation Learning SelfDoc 自我监督文档表示学习 Self-Point-Flow Self-Supervised Scene Flow Estimation From Point Clouds With Optimal 具有最优点云的自点流自监督场景流估计 Self-Supervised Video Hashing via Bidirectional Transformers 通过双向变压器的自我监督视频散列 Semantic Segmentation With Generative Models Semi-Supervised Learning and Strong Out-of-Domain Generalization生成模型的语义分割：半监督学习和强大的域外泛化 Spatial Assembly Networks for Image Representation Learning 用于图像表示学习的空间组装网络 Spatial Feature Calibration and Temporal Fusion for Effective One-Stage Video Instance Segmentation有效单阶段视频的空间特征校准和时间融合 Spherical Confidence Learning for Face Recognition 人脸识别的球形置信度学习 Surrogate Gradient Field for Latent Space Manipulation 潜在空间操作的替代梯度场 Temporal Action Segmentation From Timestamp Supervision 来自时间戳监督的时间动作分割 The Heterogeneity Hypothesis Finding Layer-Wise Differentiated Network Architectures 发现分层差异化网络架构的异质性假设 Three Birds with One Stone Multi-Task Temporal Action Detection via 三只鸟一石多任务时间动作检测 Toward Accurate and Realistic Outfits Visualization With Attention to Details 注重细节的准确和现实的服装可视化 Towards Compact CNNs via Collaborative Compression 通过协作压缩实现紧凑型 CNN Transferable Semantic Augmentation for Domain Adaptation 用于域适应的可转移语义增强 Transformation Invariant Few-Shot Object Detection 变换不变少镜头目标检测 UAV-Human A Large Benchmark for Human Behavior Understanding With Unmanned Aerial VehiclesUAV-Human：通过无人机进行人类行为理解的大基准 Uncertainty-Aware Joint Salient Object and Camouflaged Object Detection 不确定性感知联合显着目标和伪装目标检测 VirFace Enhancing Face Recognition via Unlabeled Shallow Data VirFace 通过未标记的浅层数据增强人脸识别 Virtual Fully-Connected Layer Training a Large-Scale Face Recognition Dataset With 虚拟全连接层训练大规模人脸识别数据集 Domain Adaptation With Auxiliary Target Domain-Oriented Classifier 具有辅助目标面向域分类器的域自适应 Flow-Based Kernel Prior With Application to Blind Super-Resolution 基于流的内核先验应用于盲超分辨率 High-Resolution Photorealistic Image Translation in Real-Time A Laplacian Pyramid Translation 实时高分辨率真实感图像翻译拉普拉斯金字塔翻译 OPANAS One-Shot Path Aggregation Network Architecture Search for Object Detection 用于对象检测的 OPANAS One-Shot Path Aggregation 网络架构搜索 PPR10K A Large-Scale Portrait Photo Retouching Dataset With Human-Region Mask and Group-Level ConsistencyPPR10K 具有人体区域掩码和组级一致性的大规模人像照片修饰数据集 RangeIoUDet Range Image Based Real-Time 3D Object Detector Optimized by RangeIoUDet 基于范围图像的实时 3D 对象检测器 4D Hyperspectral Photoacoustic Data Restoration With Reliability Analysis 具有可靠性分析的 4D 高光谱光声数据恢复 Image Inpainting Guided by Coherence Priors of Semantics and Textures 由语义和纹理的相干先验引导的图像修复 Towards Good Practices for Efficiently Annotating Large-Scale Image Classification Datasets 迈向有效注释大规模图像分类数据集的良好实践 Shape and Material Capture at Home 在家中捕捉形状和材料 Monocular Depth Estimation via Listwise Ranking Using the Plackett-Luce Model 使用 Plackett-Luce 模型通过 Listwise Ranking 进行单目深度估计 Building Reliable Explanations of Unreliable Neural Networks Locally Smoothing Perspective 从局部平滑视角构建不可靠神经网络的可靠解释 Anycost GANs for Interactive Image Synthesis and Editing 用于交互式图像合成和编辑的 Anycost GAN COMPLETER Incomplete Multi-View Clustering via Contrastive Prediction 通过对比预测完成不完全多视图聚类 Drafting and Revision Laplacian Pyramid Network for Fast High-Quality Artistic 用于快速高质量艺术的拉普拉斯金字塔网络的起草和修订 End-to-End Human Pose and Mesh Reconstruction with Transformers 使用变形金刚进行端到端人体姿势和网格重建 Learning Salient Boundary Feature for Anchor-free Temporal Action Localization 学习无锚时间动作定位的显着边界特征 MOOD Multi-Level Out-of-Distribution Detection MOOD 多级分布外检测 Multi-View Multi-Person 3D Pose Estimation With Plane Sweep Stereo 平面扫描立体多视图多人 3D 姿态估计 Point2Skeleton Learning Skeletal Representations from Point Clouds Point2Skeleton 从点云中学习骨骼表示 Real-Time High-Resolution Background Matting实时高分辨率背景抠图 Reciprocal Landmark Detection and Tracking With Extremely Few Annotations 具有极少注释的互惠地标检测和跟踪 Rich Context Aggregation With Reflection Prior for Glass Surface Detection 用于玻璃表面检测的具有反射先验的丰富上下文聚合 Scene-Intuitive Agent for Remote Embodied Visual Grounding 用于远程体现视觉接地的场景直观代理 Vx2Text End-to-End Learning of Video-Based Text Generation From Multimodal Inputs Vx2Text 从多模式输入中端到端学习基于视频的文本生成 What Can Style Transfer and Paintings Do for Model Robustness 样式迁移和绘画可以为模型的稳健性做些什么 AutoInt Automatic Integration for Fast Neural Volume Rendering AutoInt 用于快速神经体积渲染的自动集成 Region-Aware Adaptive Instance Normalization for Image Harmonization 用于图像协调的区域感知自适应实例归一化 3D-to-2D Distillation for Indoor Scene Parsing 用于室内场景解析的 3D 到 2D 蒸馏 Adaptive Aggregation Networks for Class-Incremental Learning 用于类增量学习的自适应聚合网络 Adaptive Cross-Modal Prototypes for Cross-Domain Visual-Language Retrieval 用于跨域视觉语言检索的自适应跨模态原型 Anti-Aliasing Semantic Reconstruction for Few-Shot Semantic Segmentation 少镜头语义分割的抗锯齿语义重建 Cluster-Wise Hierarchical Generative Model for Deep Amortized Clustering 用于深度摊销聚类的分簇层次生成模型 Content-Aware GAN Compression 内容感知 GAN 压缩 Context-Aware Biaffine Localizing Network for Temporal Sentence Grounding 用于时间句子接地的上下文感知双仿射定位网络 Cross-Modal Collaborative Representation Learning and a Large-Scale RGBT Benchmark for Crowd Counting跨模式协同表示学习和大规模 RGBT 人群计数基准 Deep Dual Consecutive Network for Human Pose Estimation 用于人体姿态估计的深度对偶连续网络 Deep Implicit Moving Least-Squares Functions for 3D Reconstruction 用于 3D 重建的深度隐式移动最小二乘函数 Deep Learning in Latent Space for Video Prediction and Compression 用于视频预测和压缩的潜在空间中的深度学习 DeepMetaHandles Learning Deformation Meta-Handles of 3D Meshes With Biharmonic Coordinates DeepMetaHandles 学习具有双调和坐标的 3D 网格的变形元句柄 DeFLOCNet Deep Image Editing via Flexible Low-Level Controls DeFLOCNet 通过灵活的低级控制进行深度图像编辑 Discovering Hidden Physics Behind Transport Dynamics 发现运输动力学背后的隐藏物理 DivCo Diverse Conditional Image Synthesis via Contrastive Generative Adversarial Network 通过对比生成对抗网络进行 DivCo 不同条件图像合成 Exploit Visual Dependency Relations for Semantic Segmentation 利用视觉依赖关系进行语义分割 Exploring and Distilling Posterior and Prior Knowledge for Radiology Report Generation探索和提炼后验和先验知识的放射学报告生成 FedDG Federated Domain Generalization on Medical Image Segmentation via Episodic FedDG 基于Episodic的医学图像分割的联邦域泛化 From Shadow Generation To Shadow Removal 从阴影生成到阴影去除 Fully Convolutional Scene Graph Generation 全卷积场景图生成 Fully Understanding Generic Objects Modeling Segmentation and Reconstruction 全面理解通用对象建模分割与重构 Generic Perceptual Loss for Modeling Structured Output Dependencies 用于建模结构化输出依赖关系的通用感知损失 Goal-Oriented Gaze Estimation for Zero-Shot Learning 零样本学习的面向目标的注视估计 iMiGUE An Identity-Free Video Dataset for Micro-Gesture Understanding and Emotion AnalysisiMiGUE 用于微手势理解和情感分析的无身份视频数据集 Inception Convolution With Efficient Dilation Search 具有高效膨胀搜索的 Inception 卷积 Invertible Denoising Network A Light Solution for Real Noise Removal 可逆去噪网络 一种用于真正去噪的轻型解决方案 Learnable Motion Coherence for Correspondence Pruning 对应剪枝的可学习运动相干性 Learning To Warp for Style Transfer 学习变形以进行风格迁移 Mask-Embedded Discriminator With Region-Based Semantic Regularization for Semi-Supervised Class-Conditional Image Synthesis用于半监督类条件图像合成的具有基于区域语义正则化的掩码嵌入鉴别器 Multimodal Motion Prediction With Stacked Transformers 堆叠变压器的多模态运动预测 Multi-Shot Temporal Event Localization A Benchmark 多镜头时间事件定位基准 Neighborhood Normalization for Robust Geometric Feature Learning 稳健几何特征学习的邻域归一化 No Frame Left Behind Full Video Action Recognition 全视频动作识别不留帧 Noise-Resistant Deep Metric Learning With Ranking-Based Instance Selection 具有基于排名的实例选择的抗噪声深度度量学习 One Thing One Click A Self-Training Approach for Weakly Supervised 一件事一点击 弱监督的自我训练方法 Orthogonal Over-Parameterized Training 正交过参数化训练 PD-GAN Probabilistic Diverse GAN for Image Inpainting 用于图像修复的 PD-GAN Probabilistic Diverse GAN PluckerNet Learn To Register 3D Line Reconstructions PluckerNet 学习注册 3D 线重建 PointGuard Provably Robust 3D Point Cloud Classification PointGuard 可证明稳健的 3D 点云分类 RankDetNet Delving Into Ranking Constraints for Object Detection RankDetNet 深入研究对象检测的排名约束 Rank-One Prior Toward Real-Time Scene Recovery 在实时场景恢复方面排名第一 Refer-It-in-RGBD A Bottom-Up Approach for 3D Visual Grounding in RGBD Refer-It-in-RGBD 一种自下而上的方法，用于 RGBD 中的 3D 视觉接地 Relation-aware Instance Refinement for Weakly Supervised Visual Grounding 弱监督视觉接地的关系感知实例细化 Retinex-Inspired Unrolling With Cooperative Prior Architecture Search for Low-Light Image EnhancementRetinex 启发的展开与协作先验架构搜索低光图像 Semi-Supervised 3D Hand-Object Poses Estimation With Interactions in Time 半监督 3D 手对象姿势估计与时间交互 SG-Net Spatial Granularity Network for One-Stage Video Instance Segmentation 用于单阶段视频实例分割的 SG-Net 空间粒度网络 Smoothing the Disentangled Latent Style Space for Unsupervised Image-to-Image Translation 平滑解开的潜在样式空间以进行无监督的图像到图像转换 Source-Free Domain Adaptation for Semantic Segmentation 语义分割的无源域自适应 Spatial-Phase Shallow Learning Rethinking Face Forgery Detection in Frequency Domain 空间阶段浅层学习重新思考频域中的人脸伪造检测 Spatial-Temporal Correlation and Topology Learning for Person Re-Identification in Videos 视频中人物重新识别的时空相关和拓扑学习 Spatiotemporal Registration for Event-Based Visual Odometry 基于事件的视觉里程计的时空配准 The Blessings of Unlabeled Background in Untrimmed Videos 未修剪视频中未标记背景的好处 Towards Unified Surgical Skill Assessment 迈向统一的手术技能评估 Unsupervised Part Segmentation Through Disentangling Appearance and Shape 通过解开外观和形状进行无监督零件分割 Watching You Global-Guided Reciprocal Learning for Video-Based Person Re-Identification 观看全球引导的互惠学习，以进行基于视频的人员重新识别 Weakly Supervised Instance Segmentation for Videos With Temporal Mask Consistency 具有时间掩模一致性的视频的弱监督实例分割 Zero-Shot Adversarial Quantization 零样本对抗量化 CLCC Contrastive Learning for Color Constancy CLCC 颜色恒常性对比学习 Multi-view Depth Estimation using Epipolar Spatio-Temporal Networks 使用对极时空网络进行多视图深