CVPR2020论文列表（中英对照）

Conditional Channel Gated Networks for Task-Aware Continual Learning 门控网络用于任务感知和持续学习的条件
Multimodal Categorization of Crisis Events in Social Media 社交媒体中危机事件的多模式分类
Counterfactual Vision and Language Learning 反事实视觉和语言学习
Gold Seeker Information Gain From Policy Distributions for Goal-Oriented Vision-and-Langauge 从目标愿景和语言的政策分配中获取寻金者信息
Image2StyleGAN How to Edit the Embedded Images Image2StyleGAN 如何编辑嵌入式图像
Cross-Modal Deep Face Normals With Deactivable Skip Connections 跨模态深度法线具有可停止跳跃连接
Hussein Correction Filter for Single Image Super-Resolution Robustifying Off-the-Shelf Deep Super-Resolvers用于单幅图像超分辨率鲁棒化现成深度超分辨率 Hussein 校正滤波器
Towards Achieving Adversarial Robustness by Enforcing Feature Consistency Across Bit 对抗性鲁棒性通过强制跨位特征的一致性来实现
Deep White-Balance Editing 深度白平衡编辑
Towards Causal VQA Revealing and Reducing Spurious Correlations by Invariant 朝着因果 VQA 揭示和减少不变量的虚假相关性
Scale-Space Flow for End-to-End Optimized Video Compression 端到端优化视频压缩的尺度空间流
Camera On-Boarding for Person Re-Identification Using Hypothesis Transfer Learning 使用假设迁移学习人员重新识别的相机载入
Density-Based Clustering for 3D Object Detection in Point Clouds 用于点云中 3D 基于密度的聚类
Non-Adversarial Video Synthesis With Learned Priors 具有学习先验的非对抗性视频合成
Fast Soft Color Segmentation 颜色分割快速柔和
From Two Rolling Shutters to One Global Shutter 从两个卷帘门到一个全局快门
Active Speakers in Context 上下文中的活跃演讲者
From Paris to Berlin Discovering Fashion Style Influences Around the 从巴黎到柏林 发现各地时尚风格的影响
Disentangled Image Generation Through Structured Noise Injection 通过结构化噪声注入生成分离的图像
A Stochastic Conditioning Scheme for Diverse Human Motion Prediction 一种用于各种人体运动预测的随机调节方案
High-Resolution Daytime Translation Without Domain Labels 日间翻译无域标签的高分辨率
A Characteristic Function Approach to Deep Implicit Generative Modeling 深度隐式生成建模的特征函数方法
Unsupervised Multi-Modal Image Registration via Geometry Preserving Image-to-Image Translation 无监督何保留图像到图像转换几何保留图像到图像转换
Single-Stage Semantic Segmentation From Image Labels 图像标签的单阶段语义分割
UniPose Unified Human Pose Estimation in Single Images and Videos 在单一图像和视频中 UniPose 估计统一的人体姿势
SAL Sign Agnostic Learning of Shapes From Raw Data 从原始数据中学习形状 SAL 不知论符号
Single-Step Adversarial Training With Dropout Scheduling 使用 Dropout 单步对抗训练的调度
TESA Tensor Element Self-Attention via Matricization 矩阵化 TESA 张量元素自我注意
Bi3D Stereo Depth Estimation via Binary Classifications 二进制分类 Bi3D 三维深度估计
Meshlet Priors for 3D Mesh Reconstruction 用于 3D 网格重建的 Meshlet 先验
Weakly-Supervised Domain Adaptation via GAN and Mesh Model for Estimating 通过 GAN 弱监督域和网格模型自适应估计
Explorable Super Resolution 超分辨率可探索
Exploring Unlabeled Faces for Novel Attribute Discovery 探索未标记的面孔，找到新的属性
Adaptive Dilated Network With Self-Correction Supervision for Counting 具有自校正监督的自适应扩张网络计数
D3Feat Joint Learning of Dense Detection and Description of 3D Local FeaturesD3Feat 密集检测和 3D 联合学习描述
Deep Facial Non-Rigid Multi-View Stereo 非刚性多视刚性多视图立体
Learning to Forget for Meta-Learning 学习忘记元学习
Event Probability Mask EPM and Event Denoising Convolutional Neural Network 事件概率掩码 EPM 和事件去噪卷积神经网络
An Adaptive Neural Network for Unsupervised Mosaic Consistency Analysis in 自适应神经网络用于无监督马赛克一致性分析
Novel Object Viewpoint Estimation Through Reconstruction Alignment 估计重建对齐的新目标
4D Visualization of Dynamic Events From Unconstrained Multi-View Videos 动态事件来自不受约束的多视图视频 4D 可视化
SAM The Sensitivity of Attribution Methods to Hyperparameters SAM 归因法对超参数的敏感性
Height and Uprightness Invariance for 3D Prediction From a Single 单次 3D 预测的高度和垂直度不变
MAGSAC a Fast Reliable and Accurate Robust Estimator MAGSAC 一种快速、可靠、准确的稳定估计器
ScopeFlow Dynamic Scene Scoping for Optical Flow ScopeFlow 用于光流动态场景范围
Improved Few-Shot Visual Classification 改进的 Few-Shot 视觉分类
Shape Reconstruction by Learning Differentiable Surface Representations 形状重建通过学习微分表面表示
Context R-CNN Long Term Temporal Context for Per-Camera Object Detection Context R-CNN Long Term Temporal Context for Per-Camera Object Detection
SpeedNet Learning the Speediness in Videos SpeedNet 学习视频中的速度
Can Weight Sharing Outperform Random Architecture Search An Investigation With 权重共享比随机架构搜索更好吗？
PandaNet Anchor-Based Single-Shot Multi-Person 3D Pose Estimation PandaNet Anchor-Based Single-Shot Multi-Person 3D Pose Estimation
Uninformed Students Student-Teacher Anomaly Detection With Discriminative Latent Embeddings 不知道学生和师生的异常检测具有判断性潜在嵌入性
AOWS Adaptive and Optimal Network Width Search With Latency Constraints 有延迟约束 AOWS 搜索自适应和最优网络宽度
MINA Convex Mixed-Integer Programmig for Non-Rigid Shape Alignment 用于非刚性形状对齐的 MINA 凸混合整数规划
Classifying Segmenting and Tracking Object Instances in Video with Mask 使用掩码对视频中的分割和跟踪对象实例进行分类
Making Better Mistakes Leveraging Class Hierarchies With Deep Networks 利用深度网络的类层次结构犯下更好的错误
DUNIT Detection-Based Unsupervised Image-to-Image Translation 基于 DUNIT 检测的无监督图像到图像转换
Normalizing Flows With Multi-Scale Autoregressive Priors 使用多尺度自回归先验规范化流
A Sparse Resultant Based Method for Efficient Minimal Solvers 一种高效最小求解器的基于稀疏结果的方法
Reinforced Feature Points Optimizing Feature Detection and Description for a 强化特征点优化特征检测和描述
Sketch Less for More On-the-Fly Fine-Grained Sketch-Based Image Retrieval 少写草图以获得更多即时基于细粒度草图的图像检索
Deep 3D Capture Geometry and Reflectance From Sparse Multi-View Images 从稀疏多视图图像中获取深度 3D 几何和反射率
ENSEI Efficient Secure Inference via Frequency-Domain Homomorphic Convolution for Privacy-Preserving ENSEI 通过频域同态卷积进行高效安全推理以保护隐私
Seeing Through Fog Without Seeing Fog Deep Multimodal Sensor Fusion 看透雾而不看雾 深度多模态传感器融合
Synchronizing Probability Measures on Rotations via Optimal Transport 通过最优传输同步旋转概率测度
Defending Against Universal Attacks Through Selective Feature Regeneration 通过选择性特征再生防御普遍攻击
Two-Shot Spatially-Varying BRDF and Shape Estimation 两次空间变化的 BRDF 和形状估计
DeepDeform Learning Non-Rigid RGB-D Reconstruction With Semi-Supervised Data DeepDeform Learning Non-Rigid RGB-D Reconstruction with Semi-Supervised Data
Learning a Neural Solver for Multiple Object Tracking 学习用于多对象跟踪的神经求解器
Rethinking Zero-Shot Video Classification End-to-End Training for Realistic Applications 重新思考针对实际应用的零镜头视频分类端到端培训
Solving Jigsaw Puzzles With Eroded Boundaries 解决具有侵蚀边界的拼图游戏
3FabRec Fast Few-Shot Face Alignment by Reconstruction 3FabRec 重建快速少镜头面对齐
Neural Head Reenactment with Latent Pose Descriptors 具有潜在姿势描述符的神经头部重演
nuScenes A Multimodal Dataset for Autonomous Driving nuScenes 用于自动驾驶的多模式数据集
Generalizing Hand Segmentation in Egocentric Videos With Uncertainty-Guided Model Adaptation 通过不确定性引导的模型自适应在以自我为中心的视频中推广手部分割
Learning a Unified Sample Weighting Network for Object Detection 学习用于目标检测的统一样本加权网络
Reconstruct Locally Localize Globally A Model Free Method for Object 重构局部局部化全局对象的无模型方法
Rethinking Differentiable Search for Mixed-Precision Neural Networks 重新思考混合精度神经网络的可微搜索
ZeroQ A Novel Zero Shot Quantization Framework ZeroQ 一种新颖的零镜头量化框架
Appearance Shock Grammar for Fast Medial Axis Extraction From Real 从实数中快速提取中轴的外观冲击文法
Sign Language Transformers Joint End-to-End Sign Language Recognition and Translation 手语变形金刚联合端到端手语识别和翻译
D2Det Towards High Quality Object Detection and Instance Segmentation D2Det 迈向高质量目标检测和实例分割
Domain Balancing Face Recognition on Long-Tailed Domains 长尾域上的域平衡人脸识别
Few-Shot Video Classification via Temporal Alignment 通过时间对齐的少镜头视频分类
Prime Sample Attention in Object Detection 目标检测中的主要样本注意
Stereoscopic Flash and No-Flash Photography for Shape and Albedo Recovery 用于形状和反照率恢复的立体闪光和无闪光摄影
Scalable Uncertainty for Computer Vision With Functional Variational Inference 具有功能变分推理的计算机视觉的可扩展不确定性
Modeling the Background for Incremental Learning in Semantic Segmentation 为语义分割中的增量学习建模背景
What It Thinks Is Important Is Important Robustness Transfers Through 它认为重要的是重要的健壮性通过
Attention-Driven Cropping for Very High Resolution Facial Landmark Detection 用于超高分辨率面部地标检测的注意力驱动裁剪
Data Uncertainty Learning in Face Recognition 人脸识别中的数据不确定性学习
Synthetic Learning Learn From Distributed Asynchronized Discriminator GAN Without Sharing 综合学习从分布式异步鉴别器 GAN 中学习，无需共享
Weakly-Supervised Semantic Segmentation via Sub-Category Exploration 通过子类别探索的弱监督语义分割
Neural Topological SLAM for Visual Navigation 用于视觉导航的神经拓扑 SLAM
JA-POLS A Moving-Camera Background Model via Joint Alignment and Partially-Overlapping JA-POLS 通过关节对齐和部分重叠的运动相机背景模型
3D Sketch-Aware Semantic Scene Completion via Semi-Supervised Structure Prior 通过半监督结构先验完成 3D 草图感知语义场景完成
A Hierarchical Graph Network for 3D Object Detection on Point 用于点上 3D 对象检测的分层图网络
A Multi-Task Mean Teacher for Semi-Supervised Shadow Detection 一种用于半监督阴影检测的多任务均值教师
A Neural Rendering Framework for Free-Viewpoint Relighting 用于自由视点重新照明的神经渲染框架
Action Segmentation With Joint Self-Supervised Temporal Domain Adaptation 联合自监督时域自适应的动作分割
Adaptive Fractional Dilated Convolution Network for Image Aesthetics Assessment 用于图像美学评估的自适应分数扩张卷积网络
AdderNet Do We Really Need Multiplications in Deep Learning AdderNet 我们真的需要深度学习中的乘法吗
Adversarial Robustness From Self-Supervised Pre-Training to Fine-Tuning 从自我监督预训练到微调的对抗鲁棒性
Auto-Tuning Structured Light by Optical Stochastic Gradient Descent 通过光学随机梯度下降自动调谐结构光
BANet Bidirectional Aggregation Network With Occlusion Handling for Panoptic Segmentation 用于全景分割的具有遮挡处理的 BANet 双向聚合网络
Better Captioning With Sequence-Level Exploration 使用序列级探索更好地字幕
BlendMask Top-Down Meets Bottom-Up for Instance Segmentation BlendMask 自上而下与自下而上的实例分割相遇
BSP-Net Generating Compact Meshes via Binary Space Partitioning BSP-Net 通过二进制空间分区生成紧凑网格
Camera Trace Erasing 相机痕迹擦除
Cops-Ref A New Dataset and Task on Compositional Referring Expression Cops-Ref 组合引用表达式的新数据集和任务
Counterfactual Samples Synthesizing for Robust Visual Question Answering 用于鲁棒视觉问答的反事实样本合成
Cross-View Tracking for Multi-Human 3D Pose Estimation at Over 100 多人 3D 姿势估计的交叉视图跟踪超过 100
Data-Efficient Semi-Supervised Learning by Reliable Edge Mining 可靠边缘挖掘的数据高效半监督学习
Domain Adaptive Image-to-Image Translation 域自适应图像到图像转换
DSGN Deep Stereo Geometry Network for 3D Object Detection 用于 3D 对象检测的 DSGN 深度立体几何网络
Dynamic Convolution Attention Over Convolution Kernels 卷积核上的动态卷积注意力
End-to-End Learnable Geometric Vision by Backpropagating PnP Optimization 通过反向传播 PnP 优化实现端到端可学习几何视觉
Fine-Grained Video-Text Retrieval With Hierarchical Graph Reasoning 具有层次图推理的细粒度视频文本检索
Frequency Domain Compact 3D Convolutional Neural Networks 频域紧凑型 3D 卷积神经网络
G2L-Net Global to Local Network for Real-Time 6D Pose Estimation 用于实时 6D 姿态估计的 G2L-Net 全局到本地网络
Harmonizing Transferability and Discriminability for Adapting Object Detectors 协调适应对象检测器的可迁移性和可辨别性
Image Search With Text Feedback by Visiolinguistic Attention Learning 视觉语言注意学习的文本反馈图像搜索
IMRAM Iterative Matching With Recurrent Attention Memory for Cross-Modal Image-Text 用于跨模态图像文本的 IMRAM 迭代匹配与循环注意记忆
Intelligent Home 3D Automatic 3D-House Design From Linguistic Descriptions Only 智能家居 3D 自动 3D 房屋设计仅来自语言描述
Label Distribution Learning on Auxiliary Label Space Graphs for Facial 面部辅助标签空间图的标签分布学习
Learning a Weakly-Supervised Video Actor-Action Segmentation Model With a Wise 使用 Wise 学习弱监督视频演员-动作分割模型
Learning Canonical Shape Space for Category-Level 6D Object Pose and 学习类别级 6D 对象姿势的规范形状空间和
Memory Enhanced Global-Local Aggregation for Video Object Detection 用于视频对象检测的内存增强全局-局部聚合
MnasFPN Learning Latency-Aware Pyramid Architecture for Object Detection on Mobile 用于移动目标检测的 MnasFPN 学习延迟感知金字塔架构
MonoPair Monocular 3D Object Detection Using Pairwise Spatial Relationships 使用成对空间关系的 MonoPair 单目 3D 对象检测
Network Adjustment Channel Search Guided by FLOPs Utilization Ratio FLOPs利用率引导的网络调整通道搜索
Norm-Aware Embedding for Efficient Person Search 用于高效人员搜索的规范感知嵌入
OASIS A Large-Scale Dataset for Single Image 3D in the OASIS 中单幅图像 3D 的大规模数据集
One-Shot Adversarial Attacks on Visual Tracking With Dual Attention 具有双重注意的视觉跟踪的一次性对抗性攻击
PuppeteerGAN Arbitrary Portrait Animation With Semantic-Aware Appearance Transformation 具有语义感知外观转换的 PuppeteerGAN 任意人像动画
Reusing Discriminators for Encoding Towards Unsupervised Image-to-Image Translation 重用鉴别器进行无监督图像到图像转换的编码
Salience-Guided Cascaded Suppression Network for Person Re-Identification 用于人员重新识别的显着性级联抑制网络
Say As You Wish Fine-Grained Control of Image Caption Generation 随心所欲 对图像标题生成进行细粒度控制
Selective Transfer With Reinforced Transfer Network for Partial Domain Adaptation 用于部分域适应的增强传输网络的选择性传输
Siamese Box Adaptive Network for Visual Tracking 用于视觉跟踪的连体框自适应网络
SLV Spatial Likelihood Voting for Weakly Supervised Object Detection 用于弱监督目标检测的 SLV 空间似然投票
State-Aware Tracker for Real-Time Video Object Segmentation 用于实时视频对象分割的状态感知跟踪器
Stochastic Sparse Subspace Clustering 随机稀疏子空间聚类
Unsupervised Learning of Intrinsic Structural Representation Points 内在结构表示点的无监督学习
CascadePSP Toward Class-Agnostic and Very High-Resolution Segmentation via Global and CascadePSP 通过全局和
Deep Stereo Using Adaptive Thin Volume Representation With Uncertainty Awareness 使用具有不确定性感知的自适应薄体积表示的深度立体声
Explaining Knowledge Distillation by Quantifying the Knowledge 通过量化知识来解释知识蒸馏
HigherHRNet Scale-Aware Representation Learning for Bottom-Up Human Pose Estimation HigherHRNet Scale-Aware Representation Learning for Bottom-Up Human Pose Estimation
Inter-Task Association Critic for Cross-Resolution Person Re-Identification Inter-Task Association Critic for Cross-Resolution Person Re-Identification
Learned Image Compression With Discretized Gaussian Mixture Likelihoods and Attention 使用离散高斯混合似然和注意力学习图像压缩
Panoptic-DeepLab A Simple Strong and Fast Baseline for Bottom-Up Panoptic Panoptic-DeepLab 自下而上 Panoptic 的简单强大且快速的基线
RiFeGAN Rich Feature Generation for Text-to-Image Synthesis From Prior Knowledge RiFeGAN 基于先验知识的文本到图像合成的丰富特征生成
Skeleton-Based Action Recognition With Shift Graph Convolutional Network 基于骨架的动作识别与移位图卷积网络
Time Flies Animating a Still Image With Time-Lapse Video As 时光飞逝 使用延时视频为静止图像制作动画
Non-Local Neural Networks With Grouped Bilinear Attentional Transforms 具有分组双线性注意变换的非局部神经网络
Implicit Functions in Feature Space for 3D Shape Reconstruction and 特征空间中的隐式函数用于 3D 形状重建和
Towards Efficient Model Compression via Learned Global Ranking 通过学习的全球排名实现高效模型压缩
Agriculture-Vision A Large Aerial Image Database for Agricultural Pattern Analysis Agriculture-Vision 用于农业模式分析的大型航空影像数据库
Assessing Image Quality Issues for Real-World Problems 评估实际问题的图像质量问题
When to Use Convolutional Neural Networks for Inverse Problems 何时使用卷积神经网络解决逆问题
Evaluating Weakly Supervised Object Localization Methods Right 正确评估弱监督对象定位方法
Cars Cant Fly Up in the Sky Improving Urban-Scene Segmentation 汽车不能在天空中飞起来改善城市场景分割
Hi-CMD Hierarchical Cross-Modality Disentanglement for Visible-Infrared Person Re-Identification 用于可见红外人员重新识别的 Hi-CMD 分层交叉模态解缠结
Scene-Adaptive Video Frame Interpolation via Meta-Learning 基于元学习的场景自适应视频帧插值
StarGAN v2 Diverse Image Synthesis for Multiple Domains StarGAN v2 多域的多样化图像合成
Task Agnostic Robust Learning on Corrupt Outputs by Correlation-Guided Mixture 通过相关引导混合对损坏输出进行任务无关的鲁棒学习
Detecting Attended Visual Targets in Video 检测视频中有人参与的视觉目标
Effectively Unbiased FID and Inception Score and Where to Find 有效无偏的 FID 和初始分数以及在哪里可以找到
Deep Non-Line-of-Sight Reconstruction 深度非视线重建
Deep Global Registration 深度全球注册
High-Dimensional Convolutional Networks for Geometric Pattern Recognition 用于几何模式识别的高维卷积网络
Learning Geocentric Object Pose in Oblique Monocular Images 学习倾斜单目图像中的地心物体姿势
P-nets Deep Polynomial Neural Networks P-nets 深度多项式神经网络
Detection in Crowded Scenes One Proposal Multiple Predictions 拥挤场景中的检测一建议多重预测
A Context-Aware Loss Function for Action Spotting in Soccer Videos 用于足球视频中动作识别的上下文感知损失函数
Bodies at Rest 3D Human Pose and Shape Estimation From 静止的身体 3D 人体姿势和形状估计
Detecting Adversarial Samples Using Influence Functions and Nearest Neighbors 使用影响函数和最近邻检测对抗样本
Editing in Style Uncovering the Local Semantics of GANs 风格编辑揭示 GAN 的局部语义
DoveNet Deep Image Harmonization via Domain Verification 通过域验证的 DoveNet 深度图像协调
Attention-Based Context Aware Reasoning for Situation Recognition 用于情境识别的基于注意力的上下文感知推理
Computing the Testing Error Without a Testing Set 在没有测试集的情况下计算测试误差
Meshed-Memory Transformer for Image Captioning 用于图像字幕的网状内存转换器
Context-Aware Human Motion Prediction 上下文感知人体运动预测
GanHand Predicting Human Grasp Affordances in Multi-Object Scenes GanHand 预测多对象场景中的人类掌握能力
Estimating Low-Rank Region Likelihood Maps 估计低秩区域似然图
Gradually Vanishing Bridge for Adversarial Domain Adaptation 对抗领域适应的逐渐消失的桥梁
Learning Dynamic Relationships for 3D Human Motion Prediction 学习用于 3D 人体运动预测的动态关系
Towards Discriminability and Diversity Batch Nuclear-Norm Maximization Under Label Insufficient 在标签不足的情况下实现可辨别性和多样性批量核范数最大化
Exploiting Joint Robustness to Adversarial Perturbations 利用联合鲁棒性应对对抗性扰动
High-Performance Long-Term Tracking With Meta-Updater 使用 Meta-Updater 进行高性能长期跟踪
Neural Point Cloud Rendering via Multi-Plane Projection 基于多平面投影的神经点云渲染
SG-NN Sparse Generative Neural Networks for Self-Supervised Scene Completion of 用于自监督场景完成的 SG-NN 稀疏生成神经网络
Probabilistic Regression for Visual Tracking 视觉跟踪的概率回归
Multi-Scale Fusion Subspace Clustering Using Similarity Constraint 使用相似性约束的多尺度融合子空间聚类
On the Detection of Digital Face Manipulation 论数字人脸处理的检测
Your Local GAN Designing Two Dimensional Local Attention Mechanisms for 你的本地 GAN 设计二维局部注意力机制
Sequential Mastery of Multiple Visual Tasks Networks Naturally Learn to 多个视觉任务的顺序掌握网络自然地学会
Lange Unsupervised Model Personalization While Preserving Privacy and Scalability An OpenLange 无监督模型个性化，同时保持隐私和可扩展性开放
RoboTHOR An Open Simulation-to-Real Embodied AI Platform RoboTHOR 一个开放的仿真到真实的体现 AI 平台
Optimal least-squares solution to the hand-eye calibration problem 手眼标定问题的最优最小二乘解
CvxNet Learnable Convex Decomposition CvxNet 可学习的凸分解
Detail-recovery Image Deraining via Context Aggregation Networks 通过上下文聚合网络进行细节恢复图像去雨
Disentangled and Controllable Face Image Generation via 3D Imitative-Contrastive Learning 通过 3D 模仿对比学习生成解开和可控的人脸图像
RetinaFace Single-Shot Multi-Level Face Localisation in the Wild 野外 RetinaFace 单次多级人脸定位
Semantic Image Manipulation Using Scene Graphs 使用场景图的语义图像处理
Guided Variational Autoencoder for Disentanglement Learning 用于解缠结学习的引导变分自动编码器
Learning Depth-Guided Convolutions for Monocular 3D Object Detection 学习用于单目 3D 目标检测的深度引导卷积
Minimal Solutions to Relative Pose Estimation From Two Views Sharing 从两个视图共享的相对姿态估计的最小解决方案
Robust Homography Estimation via Dual Principal Component Pursuit 基于双主成分追踪的鲁棒单应性估计
Learning to Observe Approximating Human Perceptual Thresholds for Detection of 学习观察近似人类感知阈值以检测
Deep Geometric Functional Maps Robust Feature Learning for Shape Correspondence 用于形状对应的深度几何功能图鲁棒特征学习
Benchmarking Adversarial Robustness on Image Classification 图像分类的对抗性鲁棒性基准测试
Bi-Directional Interaction Network for Person Search 人员搜索的双向交互网络
CentripetalNet Pursuing High-Quality Keypoint Pairs for Object Detection CentripetalNet 追求用于目标检测的高质量关键点对
Fashion Editing With Adversarial Parsing Learning 对抗性解析学习的时尚编辑
Instance Guided Proposal Network for Person Search 用于人员搜索的实例引导建议网络
Multi-Scale Boosted Dehazing Network With Dense Feature Fusion 具有密集特征融合的多尺度增强去雾网络
Robust Superpixel-Guided Attentional Adversarial Attack 强大的超像素引导注意力对抗攻击
Self-Robust 3D Point Recognition via Gather-Vector Guidance 通过聚集向量引导的自鲁棒 3D 点识别
What Can Be Transferred Unsupervised Domain Adaptation for Endoscopic Lesions 什么可以转移内窥镜病变的无监督域适应
HOPE-Net A Graph-Based Model for Hand-Object Pose Estimation HOPE-Net 一种基于图的手物体姿态估计模型
Unsupervised Magnification of Posture Deviations Across Subjects 跨受试者姿势偏差的无监督放大
The GAN That Warped Semantic Attribute Editing With Unpaired Data 使用不成对数据扭曲语义属性编辑的 GAN
Action Modifiers Learning From Adverbs in Instructional Videos 从教学视频中的副词学习动作修饰语
Associate-3Ddet Perceptual-to-Conceptual Association for 3D Point Cloud Object Detection 用于 3D 点云对象检测的 Associate-3Ddet 感知到概念关联
Correlation-Guided Attention for Corner Detection Based Visual Tracking 基于角点检测的视觉跟踪的相关引导注意
Learning Invariant Representation for Unsupervised Image Restoration 无监督图像恢复的学习不变表示
SpineNet Learning Scale-Permuted Backbone for Recognition and Localization 用于识别和定位的 SpineNet Learning Scale-Permuted Backbone
Adversarial Camouflage Hiding Physical-World Attacks With Natural Styles 用自然风格隐藏物理世界攻击的对抗伪装
Cross-Spectral Face Hallucination via Disentangling Independent Factors 通过解开独立因素的交叉光谱面部幻觉
Varicolored Image De-Hazing 杂色图像去雾
Panoptic-Based Image Synthesis 基于全景的图像合成
Vec2Face Unveil Human Faces From Their Blackbox Features in Face Vec2Face 从他们的黑盒特征中揭示人脸
Watch Your Up-Convolution CNN Based Generative Deep Neural Networks Are 观看基于上卷积 CNN 的生成深度神经网络
Learning User Representations for Open Vocabulary Image Hashtag Prediction 学习用于开放词汇图像标签预测的用户表示
Counting Out Time Class Agnostic Video Repetition Counting in the 计数时间类不可知的视频重复计数
Structured Multi-Hashing for Model Compression 用于模型压缩的结构化多重哈希
Tangent Images for Mitigating Spherical Distortion 用于减轻球面失真的切线图像
Use the Force Luke Learning to Predict Physical Forces by 使用力卢克学习预测物理力
Smooth Shells Multi-Scale Shape Registration With Functional Maps 使用功能图进行平滑壳多尺度形状配准
Uncertainty-Aware CNNs for Depth Completion Uncertainty from Beginning to End 从头到尾深度完成不确定性的不确定性感知 CNN
Fast Sparse ConvNets 快速稀疏卷积网络
Meta-Learning of Neural Architectures for Few-Shot Learning 用于少量学习的神经架构元学习
3D-MPA Multi-Proposal Aggregation for 3D Semantic Instance Segmentation 用于 3D 语义实例分割的 3D-MPA 多建议聚合
Photometric Stereo via Discrete Hypothesis-and-Test Search 通过离散假设和测试搜索的光度立体
Oops Predicting Unintentional Action in Video 糟糕，预测视频中的无意动作
A Disentangling Invertible Interpretation Network for Explaining Latent Representations 用于解释潜在表示的解缠结可逆解释网络
Learning to Discriminate Information for Online Action Detection 学习区分在线动作检测的信息
Differentiable Adaptive Computation Time for Visual Reasoning 视觉推理的可微自适应计算时间
Compressed Volumetric Heatmaps for Multi-Person 3D Pose Estimation 用于多人 3D 姿势估计的压缩体积热图
TRPLP - Trifocal Relative Pose From Lines at Points TRPLP - 点线的三焦相对位姿
Camouflaged Object Detection 伪装物体检测
Few-Shot Object Detection With Attention-RPN and Multi-Relation Detector 使用 Attention-RPN 和多关系检测器的 Few-Shot 目标检测
FGN Fully Guided Network for Few-Shot Instance Segmentation FGN 完全引导网络，用于少量实例分割
GaitPart Temporal Part-Based Model for Gait Recognition GaitPart 基于时间部分的步态识别模型
Learning Integral Objects With Intra-Class Discriminator for Weakly-Supervised Semantic Segmentation 使用类内鉴别器学习用于弱监督语义分割的整体对象
Learning Longterm Representations for Person Re-Identification Using Radio Signals 学习使用无线电信号重新识别人员的长期表示
Taking a Deeper Look at Co-Salient Object Detection 深入研究共显着目标检测
Connect-and-Slice An Hybrid Approach for Reconstructing 3D Objects Connect-and-Slice 一种用于重建 3D 对象的混合方法
Densely Connected Search Space for More Flexible Neural Architecture Search 用于更灵活的神经架构搜索的密集连接搜索空间
GraspNet-1Billion A Large-Scale Benchmark for General Object Grasping GraspNet-1Billion 通用对象抓取的大规模基准
Perceptual Quality Assessment of Smartphone Photography 智能手机摄影的感知质量评估
TPNet Trajectory Proposal Network for Motion Prediction 用于运动预测的 TPNet 轨迹提议网络
SCT Set Constrained Temporal Transformer for Set Supervised Action Segmentation 用于集监督动作分割的 SCT 集约束时间变换器
X3D Expanding Architectures for Efficient Video Recognition 用于高效视频识别的 X3D 扩展架构
Three-Dimensional Reconstruction of Human Interactions 人机交互的三维重构
ScrabbleGAN Semi-Supervised Varying Length Handwritten Text Generation ScrabbleGAN 半监督变长手写文本生成
Information-Driven Direct RGB-D Odometry 信息驱动的直接 RGB-D 里程计
How Much Time Do You Have Modeling Multi-Duration Saliency 您有多少时间建模多持续时间显着性
gDLS Generalized Pose-and-Scale Estimation Given Scale and Gravity Priors 给定尺度和重力先验的 gDLS 广义姿态和尺度估计
JL-DCF Joint Learning and Densely-Cooperative Fusion Framework for RGB-D Salient 用于 RGB-D 显着性的 JL-DCF 联合学习和密集协作融合框架
Joint Texture and Geometry Optimization for RGB-D Reconstruction RGB-D 重建的联合纹理和几何优化
MCEN Bridging Cross-Modal Gap between Cooking Recipes and Dish Images MCEN 弥合烹饪食谱和菜肴图像之间的跨模式差距
Neural Implicit Embedding for Point Cloud Analysis 用于点云分析的神经隐式嵌入
Learning Generative Models of Shape Handles 学习形状手柄的生成模型
Wish You Were Here Context-Aware Human Generation 希望你在这里 具有上下文意识的人类一代
Music Gesture for Visual Sound Separation 用于视觉声音分离的音乐手势
AdversarialNAS Adversarial Neural Architecture Search for GANs AdversarialNAS 对抗性神经架构搜索 GAN
Discrete Model Compression With Resource Constraint for Deep Neural Networks 具有资源约束的深度神经网络离散模型压缩
Flow Contrastive Estimation of Energy-Based Models 基于能量的模型的流动对比估计
GraphTER Unsupervised Learning of Graph Transformation Equivariant Representations via Auto-Encoding GraphTER 通过自动编码对图变换等变表示进行无监督学习
Learning to Optimize on SPD Manifolds 学习优化 SPD 歧管
Listen to Look Action Recognition by Previewing Audio 通过预览音频收听 Look 动作识别
MTL-NAS Task-Agnostic Neural Architecture Search Towards General-Purpose Multi-Task Learning 面向通用多任务学习的 MTL-NAS 任务不可知神经架构搜索
Multi-Modal Graph Neural Network for Joint Reasoning on Vision and 用于视觉和视觉联合推理的多模态图神经网络
Pose-Guided Visible Part Matching for Occluded Person ReID 被遮挡人员 ReID 的姿势引导可见部分匹配
Recursive Least-Squares Estimator-Aided Online Learning for Visual Tracking 用于视觉跟踪的递归最小二乘估计器辅助在线学习
SketchyCOCO Image Generation From Freehand Scene Sketches 从手绘场景草图生成 SketchyCOCO 图像
VectorNet Encoding HD Maps and Agent Dynamics From Vectorized Representation VectorNet 从矢量化表示编码高清地图和代理动态
Satellite Image Time Series Classification With Pixel-Set Encoders and Temporal 使用像素集编码器和时间序列进行卫星图像时间序列分类
Actor-Transformers for Group Activity Recognition 用于群体活动识别的 Actor-Transformers
Video to Events Recycling Video Datasets for Event Cameras 视频到事件 为事件摄像机回收视频数据集
Averaging Essential and Fundamental Matrices in Collinear Camera Settings 在共线相机设置中平均基本矩阵和基本矩阵
Local Deep Implicit Functions for 3D Shape 3D 形状的局部深层隐式函数
Learning Representations by Predicting Bags of Visual Words 通过预测视觉词袋来学习表示
Learning Multiview 3D Point Cloud Registration 学习多视图 3D 点云配准
Eternal Sunshine of the Spotless Net Selective Forgetting in Deep 一尘不染的永恒阳光 选择性遗忘在深处
ReSprop Reuse Sparsified Backpropagation ReSprop 重用稀疏反向传播
A Quantum Computational Approach to Correspondence Problems on Point Sets 点集对应问题的一种量子计算方法
Geometrically Principled Connections in Graph Neural Networks 图神经网络中的几何原理连接
Learning Temporal Co-Attention Models for Unsupervised Video Action Localization 学习用于无监督视频动作定位的时间共同注意模型
Achieving Robustness in the Wild via Adversarial Mixing With Disentangled 通过对抗性混合与 Disentangled 实现野外鲁棒性
Dynamic Neural Relational Inference 动态神经关系推理
Cascade Cost Volume for High-Resolution Multi-View Stereo and Stereo Matching 高分辨率多视图立体和立体匹配的级联成本量
Image Processing Using Multi-Code GAN Prior 使用多码 GAN 先验的图像处理
Improving the Robustness of Capsule Networks to Image Affine Transformations 提高胶囊网络对图像仿射变换的鲁棒性
Spherical Space Domain Adaptation With Robust Pseudo-Label Loss 具有鲁棒伪标签损失的球面空间域自适应
Generative Hybrid Representations for Activity Forecasting With No-Regret Learning 无悔学习活动预测的生成混合表示
Minimal Solutions for Relative Pose With a Single Affine Correspondence 具有单个仿射对应的相对位姿的最小解
Through Fog High-Resolution Imaging Using Millimeter Wave Radar 使用毫米波雷达通过雾高分辨率成像
Deep Active Learning for Biased Datasets via Fisher Kernel Self-Supervision 通过 Fisher 内核自我监督的有偏数据集的深度主动学习
FeatureFlow Robust Video Interpolation via Structure-to-Texture Generation 通过结构到纹理生成的 FeatureFlow 鲁棒视频插值
3D Packing for Self-Supervised Monocular Depth Estimation 用于自监督单目深度估计的 3D 打包
A Spatiotemporal Volumetric Interpolation Network for 4D Dynamic Medical Image 一种用于 4D 动态医学图像的时空体积插值网络
Attentive Weights Generation for Few Shot Learning via Information Maximization 通过信息最大化为少数镜头学习生成注意力权重
AugFPN Improving Multi-Scale Feature Learning for Object Detection AugFPN 改进用于目标检测的多尺度特征学习
Closed-Loop Matters Dual Regression Networks for Single Image Super-Resolution 闭环对单图像超分辨率的双回归网络很重要
Density-Aware Feature Embedding for Face Clustering 用于人脸聚类的密度感知特征嵌入
DMCP Differentiable Markov Channel Pruning for Neural Networks 神经网络的 DMCP 可微马尔可夫通道修剪
Hit-Detector Hierarchical Trinity Architecture Search for Object Detection Hit-Detector Hierarchical Trinity Architecture Search for Object Detection
Iterative Context-Aware Graph Inference for Visual Dialog 视觉对话的迭代上下文感知图推理
Learning Meta Face Recognition in Unseen Domains 在未知领域学习元人脸识别
Multi-Dimensional Pruning A Unified Framework for Model Compression 多维剪枝模型压缩的统一框架
Normalized and Geometry-Aware Self-Attention Network for Image Captioning 用于图像描述的归一化和几何感知自注意力网络
On Positive-Unlabeled Classification in GAN 关于 GAN 中的正无标签分类
Online Knowledge Distillation via Collaborative Learning 通过协作学习进行在线知识提炼
Organ at Risk Segmentation for Head and Neck Cancer Using 头颈癌的高危器官分割使用
SiamCAR Siamese Fully Convolutional Classification and Regression for Visual Tracking 用于视觉跟踪的 SiamCAR Siamese 全卷积分类和回归
When NAS Meets Robustness In Search of Robust Architectures Against 当 NAS 在寻找稳健架构时遇到稳健性
Zero-Reference Deep Curve Estimation for Low-Light Image Enhancement 低光图像增强的零参考深度曲线估计
PatchVAE Learning Local Latent Codes for Recognition PatchVAE 学习本地潜在代码进行识别
Rethinking Depthwise Separable Convolutions How Intra-Kernel Correlations Lead to Improved 重新思考深度可分离卷积内核内相关性如何导致改进
DeepCap Monocular Human Performance Capture Using Weak Supervision 使用弱监督的 DeepCap 单目人体性能捕获
HOnnotate A Method for 3D Annotation of Hand and Object HOnnotate 一种对手和物体进行 3D 标注的方法
GhostNet More Features From Cheap Operations GhostNet 来自廉价操作的更多功能
Joint Training of Variational Auto-Encoder and Latent Energy-Based Model 变分自编码器与基于潜在能量的模型的联合训练
Learning the Redundancy-Free Features for Generalized Zero-Shot Object Recognition 学习用于广义零样本目标识别的无冗余特征
Neuromorphic Camera Guided High Dynamic Range Imaging 神经形态摄像机引导的高动态范围成像
OccuSeg Occupancy-Aware 3D Instance Segmentation OccuSeg 占用感知 3D 实例分割
RMP-SNN Residual Membrane Potential Neuron for Enabling Deeper High-Accuracy and RMP-SNN 残膜电位神经元，用于实现更深层次的高精度和
SPARE3D A Dataset for SPAtial REasoning on Three-View Line Drawings SPARE3D 用于三视图线图空间推理的数据集
DualSDF Semantic Shape Manipulation Using a Two-Level Representation 使用两级表示的 DualSDF 语义形状操作
Towards Learning a Generic Agent for Vision-and-Language Navigation via Pre-Training 通过预训练学习视觉和语言导航的通用代理
ILFO Adversarial Attack on Adaptive Neural Networks ILFO 对自适应神经网络的对抗性攻击
Space-Time-Aware Multi-Resolution Video Enhancement 时空感知多分辨率视频增强
The Knowledge Within Methods for Data-Free Model Compression 无数据模型压缩方法中的知识
Multi-scale Domain-adversarial Multiple-instance CNN for Cancer Subtype Classification with Unannotated 用于未注释癌症亚型分类的多尺度域对抗多实例 CNN
Leveraging Photometric Consistency Over Time for Sparsely Supervised Hand-Object Reconstruction 利用光度一致性随着时间的推移进行稀疏监督的手对象重建
MPM Joint Representation of Motion and Position Map for Cell 细胞运动和位置图的 MPM 联合表示
Nonparametric Object and Parts Modeling With Lie Group Dynamics 使用李群动力学的非参数对象和零件建模
Defending and Harnessing the Bit-Flip Based Adversarial Weight Attack 防御和利用基于位翻转的对抗性权重攻击
Epipolar Transformers 对极变压器
Incremental Learning in Online Scenario 在线场景中的增量学习
Learning Filter Pruning Criteria for Deep Convolutional Neural Networks Acceleration 深度卷积神经网络加速的学习过滤器修剪标准
MiLeNAS Efficient Neural Architecture Search via Mixed-Level Reformulation 通过混合级重构的 MiLeNAS 高效神经架构搜索
Momentum Contrast for Unsupervised Visual Representation Learning 无监督视觉表示学习的动量对比
PVN3D A Deep Point-Wise 3D Keypoints Voting Network for 6DoF PVN3D 用于 6DoF 的深度 Point-Wise 3D 关键点投票网络
Structure Aware Single-Stage 3D Object Detection From Point Cloud 基于点云的结构感知单阶段 3D 对象检测
A Lighting-Invariant Point Processor for Shading 用于着色的光照不变点处理器
Leveraging 2D Data to Learn Textured 3D Mesh Generation 利用 2D 数据学习带纹理的 3D 网格生成
Learning a Neural 3D Texture Space From 2D Exemplars 从 2D 样本中学习神经 3D 纹理空间
A Multi-Hypothesis Approach to Color Constancy 颜色恒定性的多假设方法
Learning to Autofocus 学习自动对焦
PointGMM A Neural GMM Network for Point Clouds PointGMM 用于点云的神经 GMM 网络
Exploit Clues From Views Self-Supervised and Regularized Learning for Multiview 利用视图中的线索进行多视图的自我监督和正则化学习
EPOS Estimating 6D Pose of Objects With Symmetries EPOS 估计具有对称性的物体的 6D 位姿
Augment Your Batch Improving Generalization Through Instance Repetition 通过实例重复增强您的批次提高泛化能力
Distilling Image Dehazing With Heterogeneous Task Imitation 使用异构任务模仿蒸馏图像去雾
Learning to Detect Important People in Unlabelled Images for Semi-Supervised 学习在半监督的未标记图像中检测重要人物
Composed Query Image Retrieval Using Locally Bounded Features 使用局部有界特征的组合查询图像检索
Inter-Region Affinity Distillation for Road Marking Segmentation 用于道路标记分割的区域间亲和蒸馏
Learning to Structure an Image With Few Colors 学习用少量颜色构建图像
Real-Time Panoptic Segmentation From Dense Detections 密集检测的实时全景分割
RevealNet Seeing Behind Objects in RGB-D Scans RevealNet 在 RGB-D 扫描中看到物体背后
Strip Pooling Rethinking Spatial Pooling for Scene Parsing Strip Pooling 重新思考用于场景解析的空间池化
ViBE Dressing for Diverse Body Shapes 适合不同体型的 ViBE 敷料
Generalized ODIN Detecting Out-of-Distribution Image Without Learning From Out-of-Distribution Data 广义 ODIN 检测分布外图像而不从分布外数据中学习
Bi-Directional Relationship Inferring Network for Referring Image Segmentation 用于参考图像分割的双向关系推断网络
Collaborative Motion Prediction via Neural Motion Message Passing 通过神经运动消息传递的协同运动预测
Creating Something From Nothing Unsupervised Knowledge Distillation for Cross-Modal Hashing 从无到有的无监督知识蒸馏用于跨模态散列
DSNAS Direct Neural Architecture Search Without Parameter Retraining 无参数再训练的 DSNAS 直接神经架构搜索
Iterative Answer Prediction With Pointer-Augmented Multimodal Transformers for TextVQA 用于 TextVQA 的指针增强多模态变换器的迭代答案预测
Learning to Segment the Tail 学习分割尾巴
Progressive Relation Learning for Group Activity Recognition 用于群体活动识别的渐进式关系学习
RandLA-Net Efficient Semantic Segmentation of Large-Scale Point Clouds 大规模点云的 RandLA-Net 高效语义分割
Single-Stage 6D Object Pose Estimation 单阶段 6D 物体姿态估计
Temporally Distributed Networks for Fast Video Semantic Segmentation 用于快速视频语义分割的时间分布式网络
Unsupervised Domain Adaptation With Hierarchical Gradient Synchronization 具有分层梯度同步的无监督域自适应
What You See is What You Get Exploiting Visibility for 所见即所得
Adversarial Texture Optimization From RGB-D Scans 来自 RGB-D 扫描的对抗性纹理优化
An Internal Covariate Shift Bounding Algorithm for Deep Neural Networks 一种用于深度神经网络的内部协变量移位边界算法
An Investigation Into the Stochasticity of Batch Whitening 批量美白随机性调查
ARCH Animatable Reconstruction of Clothed Humans ARCH 穿衣人的动画重建
ClusterVO Clustering Moving Instances and Estimating Visual Odometry for Self ClusterVO 聚类移动实例和估计自身的视觉里程计
Controllable Orthogonalization in Training DNNs 训练 DNN 中的可控正交化
CurricularFace Adaptive Curriculum Learning Loss for Deep Face Recognition 用于深度人脸识别的 CurricularFace 自适应课程学习损失
Deep Semantic Clustering by Partition Confidence Maximisation 分区置信度最大化的深度语义聚类
Fast Video Object Segmentation With Temporal Aggregation Network and Dynamic 使用时间聚合网络和动态的快速视频对象分割
Feature-Metric Registration A Fast Semi-Supervised Approach for Robust Point Cloud 特征度量注册一种快速的半监督鲁棒点云方法
Improving Action Segmentation via Graph-Based Temporal Reasoning 通过基于图的时间推理改进动作分割
Interpretable and Accurate Fine-grained Recognition via Region Grouping 通过区域分组进行可解释且准确的细粒度识别
Learning Identity-Invariant Motion Representations for Cross-ID Face Reenactment 学习用于跨 ID 人脸重演的身份不变运动表示
NMS by Representative Region Towards Crowded Pedestrian Detection by Proposal NMS by Representative Region to Crowded Pedestrian Detection by Proposal
OctSqueeze Octree-Structured Entropy Model for LiDAR Compression 用于 LiDAR 压缩的 OctSqueeze 八叉树结构熵模型
PF-Net Point Fractal Network for 3D Point Cloud Completion PF-Net Point Fractal Network for 3D Point Cloud Completion
Probability Weighted Compact Feature for Domain Adaptive Retrieval 域自适应检索的概率加权紧凑特征
PropagationNet Propagate Points to Curve to Learn Structure Information PropagationNet 将点传播到曲线以学习结构信息
Real-World Person Re-Identification via Degradation Invariance Learning 通过退化不变学习重新识别真实世界的人
Referring Image Segmentation via Cross-Modal Progressive Comprehension 通过跨模态渐进理解引用图像分割
SQE a Self Quality Evaluation Metric for Parameters Optimization in SQE 参数优化的自我质量评估指标
The Devil Is in the Details Delving Into Unbiased Data 魔鬼在细节中钻研无偏数据
Universal Physical Camouflage Attacks on Object Detectors 对物体探测器的通用物理伪装攻击
Self-Supervised Monocular Scene Flow Estimation 自监督单目场景流估计
A Shared Multi-Attention Framework for Multi-Label Zero-Shot Learning 用于多标签零样本学习的共享多注意框架
Fine-Grained Generalized Zero-Shot Learning via Dense Attribute-Based Attention 通过基于密集属性的注意力的细粒度广义零样本学习
Interactive Multi-Label CNN Learning With Partial Labels 带有部分标签的交互式多标签 CNN 学习
Learning to Super Resolve Intensity Images From Events 从事件中学习超分辨率强度图像
Semi-Supervised Semantic Image Segmentation With Self-Correcting Networks 具有自校正网络的半监督语义图像分割
Low-Rank Compression of Neural Nets Learning the Rank of Each 学习每个等级的神经网络的低等级压缩
Global Optimality for Point Set Registration Using Semidefinite Programming 使用半定规划的点集配准的全局最优性
Weakly-Supervised 3D Human Pose Learning via Multi-View Images in the 基于多视图图像的弱监督 3D 人体姿态学习
Enhancing Generic Segmentation With Learned Region Representations 使用学习的区域表示增强通用分割
DOA-GAN Dual-Order Attentive Generative Adversarial Network for Image Copy-Move Forgery 用于图像复制移动伪造的 DOA-GAN 双阶注意力生成对抗网络
Video Super-Resolution With Temporal Group Attention 具有时间组注意的视频超分辨率
Optical Non-Line-of-Sight Physics-Based 3D Human Pose Estimation 基于光学非视线物理的 3D 人体姿态估计
Scene Recomposition by Learning-Based ICP 基于学习的ICP场景重构
Can Deep Learning Recognize Subtle Human Activities 深度学习能否识别微妙的人类活动
ActionBytes Learning From Trimmed Videos to Localize Actions ActionBytes 从修剪过的视频中学习以本地化操作
Self-Supervised Learning of Interpretable Keypoints From Unlabelled Videos 未标记视频中可解释关键点的自监督学习
Attack to Explain Deep Representation 攻击解释深度表示
Rethinking Class-Balanced Methods for Long-Tailed Visual Recognition From a Domain 重新思考领域中长尾视觉识别的类平衡方法
Generalized Product Quantization Network for Semi-Supervised Image Retrieval 用于半监督图像检索的广义乘积量化网络
xMUDA Cross-Modal Unsupervised Domain Adaptation for 3D Semantic Segmentation xMUDA 用于 3D 语义分割的跨模态无监督域自适应
Learn2Perturb An End-to-End Feature Perturbation Learning to Improve Adversarial Robustness Learn2Perturb 一种端到端的特征扰动学习来提高对抗性鲁棒性
Steering Self-Supervised Feature Learning Beyond Local Pixel Statistics 引导超越局部像素统计的自我监督特征学习
Sparse Layered Graphs for Multi-Object Segmentation 用于多对象分割的稀疏分层图
Action Genome Actions As Compositions of Spatio-Temporal Scene Graphs 动作基因组动作作为时空场景图的组合
Attention Convolutional Binary Neural Tree for Fine-Grained Visual Categorization 用于细粒度视觉分类的注意力卷积二元神经树
Revisiting Saliency Metrics Farthest-Neighbor Area Under Curve 重新审视曲线下的显着性度量最邻近区域
Single-Side Domain Generalization for Face Anti-Spoofing 人脸反欺骗的单边域泛化
Attention Scaling for Crowd Counting 人群计数的注意力缩放
Coherent Reconstruction of Multiple Humans From a Single Image 从单个图像中对多个人进行相干重建
DeeperForensics-1.0 A Large-Scale Dataset for Real-World Face Forgery Detection DeeperForensics-1.0 用于真实世界人脸伪造检测的大规模数据集
End-to-End 3D Point Cloud Instance Segmentation Without Detection 无需检测的端到端 3D 点云实例分割
Fantastic Answers and Where to Find Them Immersive Question-Directed Visual 奇妙的答案以及在哪里可以找到它们 沉浸式问题导向的视觉
In Defense of Grid Features for Visual Question Answering 捍卫视觉问答的网格特征
Learning Event-Based Motion Deblurring 学习基于事件的运动去模糊
Local Implicit Grid Representations for 3D Scenes 3D 场景的局部隐式网格表示
Multi-Scale Progressive Fusion Network for Single Image Deraining 用于单幅图像去雨的多尺度渐进融合网络
Peek-a-Boo Occlusion Reasoning in Indoor Scenes With Plane Representations 具有平面表示的室内场景中的 Peek-a-Boo 遮挡推理
PointGroup Dual-Set Point Grouping for 3D Instance Segmentation 用于 3D 实例分割的 PointGroup 双设定点分组
PSGAN Pose and Expression Robust Spatial-Aware GAN for Customizable Makeup PSGAN Pose and Expression Robust Spatial-Aware GAN 用于可定制的化妆
SDFDiff Differentiable Rendering of Signed Distance Fields for 3D Shape 3D 形状的带符号距离场的 SDFDiff 可微分渲染
SP-NAS Serial-to-Parallel Backbone Search for Object Detection SP-NAS 用于目标检测的串行到并行骨干搜索
AdaBits Neural Network Quantization With Adaptive Bit-Widths 具有自适应位宽的 AdaBits 神经网络量化
Exploring Spatial-Temporal Multi-Frequency Analysis for High-Fidelity and Temporal-Consistency Video Prediction 探索高保真和时间一致性视频预测的时空多频分析
Geometric Structure Based and Regularized Depth Estimation From 360 Indoor 360度室内基于几何结构的正则化深度估计
Light Field Spatial Super-Resolution via Deep Combinatorial Geometry Embedding and 基于深度组合几何嵌入的光场空间超分辨率和
Style Normalization and Restitution for Generalizable Person Re-Identification 可泛化人物重新识别的风格规范化和恢复
Cross-Modal Cross-Domain Moment Alignment Network for Person Search 用于人员搜索的跨模式跨域矩对齐网络
Self-Supervised Monocular Trained Depth Estimation Using Self-Attention and Discrete Disparity 使用自我注意和离散视差的自我监督单目训练深度估计
Select to Better Learn Fast and Accurate Deep Learning Using 选择以更好地学习快速准确的深度学习使用
Cylindrical Convolutional Networks for Joint Object Detection and Viewpoint Estimation 用于联合目标检测和视点估计的圆柱卷积网络
MMTM Multimodal Transfer Module for CNN Fusion 用于 CNN 融合的 MMTM 多模态传输模块
Deep Polarization Cues for Transparent Object Segmentation 用于透明对象分割的深度极化线索
Benchmarking the Robustness of Semantic Segmentation Models 对语义分割模型的鲁棒性进行基准测试
Noise Robust Generative Adversarial Networks 噪声鲁棒生成对抗网络
Defending Against Model Stealing Attacks With Adaptive Misinformation 使用自适应错误信息防御模型窃取攻击
MSG-GAN Multi-Scale Gradients for Generative Adversarial Networks 用于生成对抗网络的 MSG-GAN 多尺度梯度
Analyzing and Improving the Image Quality of StyleGAN StyleGAN 图像质量分析与改进
Deblurring Using Analysis-Synthesis Networks Pair 使用分析-合成网络对去模糊
On Translation Invariance in CNNs Convolutional Layers Can Exploit Absolute 关于 CNN 中的平移不变性卷积层可以利用 Absolute
Multiple Anchor Learning for Visual Object Detection 视觉对象检测的多锚学习
RGBD-Dog Predicting Canine Pose from RGBD Sensors RGBD-Dog 从 RGBD 传感器预测犬类姿势
RankMI A Mutual Information Maximizing Ranking Loss RankMI A 互信息最大化排名损失
Generalized Zero-Shot Learning via Over-Complete Distribution 通过过完全分布的广义零样本学习
AnimalWeb A Large-Scale Hierarchical Dataset of Annotated Animal Faces AnimalWeb 带注释动物面孔的大规模分层数据集
Hyperbolic Image Embeddings 双曲线图像嵌入
ActiveMoCap Optimized Viewpoint Selection for Active Human Motion Capture 主动人体运动捕捉的 ActiveMoCap 优化视点选择
A Programmatic and Semantic Approach to Explaining and Debugging Neural 一种解释和调试神经网络的程序化和语义化方法
Advisable Learning for Self-Driving Vehicles by Internalizing Observation-to-Action Rules 通过内化观察到行动的规则为自动驾驶汽车提供可取的学习
GroupFace Learning Latent Groups and Constructing Group-Based Representations for Face GroupFace 学习潜在组和构建基于组的人脸表示
Hypergraph Attention Networks for Multimodal Learning 用于多模态学习的超图注意网络
Learning Texture Invariant Representation for Domain Adaptation of Semantic Segmentation 语义分割领域适应的学习纹理不变表示
Learning to Simulate Dynamic Environments With GameGAN 学习使用 GameGAN 模拟动态环境
M2m Imbalanced Classification via Major-to-Minor Translation 通过从大到小翻译的 M2m 不平衡分类
Modality Shifting Attention Network for Multi-Modal Video Question Answering 用于多模态视频问答的模态转移注意力网络
Modeling Biological Immunity to Adversarial Examples 对对抗样本的生物免疫建模
Proxy Anchor Loss for Deep Metric Learning 深度度量学习的代理锚损失
Regularization on Spatio-Temporally Smoothed Feature for Action Recognition 动作识别时空平滑特征的正则化
Single Image Reflection Removal With Physically-Based Training Images 使用基于物理的训练图像去除单图像反射
Spatially Attentive Output Layer for Image Classification 用于图像分类的空间注意输出层
Transfer Learning From Synthetic to Real-Noise Denoising With Adaptive Instance 使用自适应实例将学习从合成迁移到真实噪声去噪
Video Panoptic Segmentation 视频全景分割4700，全景分割10000
PointRend Image Segmentation As Rendering PointRend 图像分割作为渲染
CONSAC Robust Multi-Model Fitting by Conditional Sample Consensus 基于条件样本一致性的 CONSAC 稳健多模型拟合
Belief Propagation Reloaded Learning BP-Layers for Labeling Problems Belief Propagation Reloaded Learning BP-Layers for Labeling Problems
Embedding Expansion Augmentation in Embedding Space for Deep Metric Learning 在深度度量学习的嵌入空间中嵌入扩展增强
Total Deep Variation for Linear Inverse Problems 线性逆问题的总深度变差
VIBE Video Inference for Human Body Pose and Shape Estimation 用于人体姿势和形状估计的 VIBE 视频推理
Universal Litmus Patterns Revealing Backdoor Attacks in CNNs 揭示 CNN 中后门攻击的通用试金石模式
PhysGAN Generating Physical-World-Resilient Adversarial Examples for Autonomous Driving PhysGAN 为自动驾驶生成物理世界弹性对抗示例
Compositional Convolutional Neural Networks A Deep Architecture With Innate Robustness 组合卷积神经网络具有内在鲁棒性的深层架构
Factorized Higher-Order CNNs With an Application to Spatio-Temporal Emotion Estimation 用于时空情绪估计的分解高阶 CNN
DeepFaceFlow In-the-Wild Dense 3D Facial Motion Estimation DeepFaceFlow In-the-Wild 密集 3D 面部运动估计
Learning Interactions and Relationships Between Movie Characters 学习电影角色之间的互动和关系
Instance Segmentation of Biological Images Using Harmonic Embeddings 使用谐波嵌入的生物图像实例分割
Articulation-Aware Canonical Surface Mapping 关节感知规范曲面映射
Weakly-Supervised Mesh-Convolutional Hand Reconstruction in the Wild 弱监督网格卷积手部重建
LUVLi Face Alignment Estimating Landmarks Location Uncertainty and Visibility Likelihood LUVLi 人脸对齐估计地标位置不确定性和可见性可能性
Self-Supervised 3D Human Pose Estimation via Part Guided Novel Image 基于部分引导新图像的自我监督 3D 人体姿态估计
Towards Inheritable Models for Open-Set Domain Adaptation 面向开放集域适应的可继承模型
Universal Source-Free Domain Adaptation 通用无源域适配
Normal Assisted Stereo Depth Estimation 正常辅助立体深度估计
Structured Compression by Weight Encryption for Unstructured Pruning and Quantization 用于非结构化剪枝和量化的权重加密结构化压缩
Blur Aware Calibration of Multi-Focus Plenoptic Camera 多焦点全光相机的模糊感知校准
Prior Guided GAN Based Semantic Inpainting 先前引导的基于 GAN 的语义修复
MAST A Memory-Augmented Self-Supervised Tracker MAST 一种记忆增强的自我监督跟踪器
MSeg A Composite Dataset for Multi-Domain Semantic Segmentation MSeg 用于多域语义分割的复合数据集
SaccadeNet A Fast and Accurate Object Detector SaccadeNet 一种快速准确的目标检测器
SampleNet Differentiable Point Cloud Sampling SampleNet 可微分点云采样
Which Is Plagiarism Fashion Image Retrieval Based on Regional Representation 基于区域表征的抄袭时尚图像检索是什么
AvatarMe Realistically Renderable 3D Facial Reconstruction In-the-Wild AvatarMe 真实可渲染的 3D 面部重建在野外
Learning Instance Occlusion for Panoptic Segmentation 全景分割的学习实例遮挡
A Graduated Filter Method for Large Scale Robust Estimation 一种用于大规模鲁棒估计的分级滤波方法
Deep Homography Estimation for Dynamic Scenes 动态场景的深度单应性估计
Going Deeper With Lean Point Networks 使用精益点网络更深入
Guen Disentangling Physical Dynamics From Unknown Factors for Unsupervised Video PredictionGuen 将物理动力学从未知因素中解脱出来，用于无监督视频预测
Hierarchical Conditional Relation Networks for Video Question Answering 用于视频问答的分层条件关系网络
AdaCoF Adaptive Collaboration of Flows for Video Frame Interpolation 视频帧插值流的 AdaCoF 自适应协作
Adversarial Vertex Mixup Toward Better Adversarially Robust Generalization 对抗性顶点混合以实现更好的对抗性鲁棒泛化
CenterMask Real-Time Anchor-Free Instance Segmentation CenterMask 实时无锚实例分割
Continual Learning With Extended Kronecker-Factored Approximate Curvature 扩展克罗内克因子近似曲率的持续学习
Large Scale Video Representation Learning via Relational Graph Clustering 基于关系图聚类的大规模视频表示学习
Learning Augmentation Network via Influence Functions 通过影响函数学习增强网络
MaskGAN Towards Diverse and Interactive Facial Image Manipulation MaskGAN 迈向多样化和交互式面部图像处理
NeuralScale Efficient Scaling of Neurons for Resource-Constrained Deep Neural Networks NeuralScale 用于资源受限的深度神经网络的神经元的有效缩放
Reference-Based Sketch Image Colorization Using Augmented-Self Reference and Dense Semantic 使用增强自我参考和密集语义的基于参考的草图图像着色
Structure Boundary Preserving Segmentation for Medical Image With Ambiguous Boundary 具有模糊边界的医学图像的结构保边界分割
TextureFusion High-Quality Texture Acquisition for Real-Time RGB-D Scanning TextureFusion 用于实时 RGB-D 扫描的高质量纹理采集
Uncertainty-Aware Mesh Decoder for High Fidelity 3D Face Reconstruction 用于高保真 3D 人脸重建的不确定性感知网格解码器
Warping Residual Based Image Stitching for Large Parallax 基于翘