计算机视觉最新论文分享 2023.3.31 | 您所在的位置:网站首页 › 微光图像数据集 › 计算机视觉最新论文分享 2023.3.31 |
计算机视觉论文分享 共计217篇 3D|Video|Temporal|Action|Multi-view相关(42篇)[1] DRIMET: Deep Registration for 3D Incompressible Motion Estimation in Tagged-MRI with Application to the Tongue 标题:DRIMET:标记MRI中3D不可压缩运动估计的深度配准及其在舌头上的应用 链接:https://arxiv.org/abs/2301.07234 发表或投稿:无 代码:未开源 [2] Self-Supervised Video Forensics by Audio-Visual Anomaly Detection 标题:基于视听异常检测的自监督视频取证 链接:https://arxiv.org/abs/2301.01767 发表或投稿:CVPR 代码:https://cfeng16.github.io/audio-visual-forensics [3] Multi-View Azimuth Stereo via Tangent Space Consistency 标题:通过切线空间一致性实现多视图方位立体 链接:https://arxiv.org/abs/2303.16447 发表或投稿:CVPR 代码:https://xucao-42.github.io/mvas_homepage/ [4] Unlocking Masked Autoencoders as Loss Function for Image and Video Restoration 标题:解锁掩码自动编码器作为图像和视频恢复的损失函数 链接:https://arxiv.org/abs/2303.16411 发表或投稿:无 代码:未开源 [5] TimeBalance: Temporally-Invariant and Temporally-Distinctive Video Representations for Semi-Supervised Action Recognition 标题:时间平衡:用于半监督动作识别的时间不变和时间区分视频表示 链接:https://arxiv.org/abs/2303.16268 发表或投稿:CVPR 代码:https://github.com/DAVEISHAN/TimeBalance [6] SparseNeRF: Distilling Depth Ranking for Few-shot Novel View Synthesis 标题:SparseNeRF:少镜头新颖视角合成的提取深度排名 链接:https://arxiv.org/abs/2303.16196 发表或投稿:无 代码:https://sparsenerf.github.io/. [7] VMesh: Hybrid Volume-Mesh Representation for Efficient View Synthesis 标题:VMesh:用于高效视图合成的混合体网格表示 链接:https://arxiv.org/abs/2303.16184 发表或投稿:无 代码:https://bennyguo.github.io/vmesh/ [8] One-Stage 3D Whole-Body Mesh Recovery with Component Aware Transformer 标题:使用组件感知转换器的一阶段3D全身网格恢复 链接:https://arxiv.org/abs/2303.16160 发表或投稿:CVPR 代码:https://osx-ubody.github.io/ [9] VIDIMU. Multimodal video and IMU kinematic dataset on daily life activities using affordable devices 标题:维迪穆。使用价格合理的设备进行日常生活活动的多模式视频和IMU运动学数据集 链接:https://arxiv.org/abs/2303.16150 发表或投稿:无 代码:未开源 [10] CycleACR: Cycle Modeling of Actor-Context Relations for Video Action Detection 标题:CycleACR:用于视频动作检测的演员-上下文关系的循环建模 链接:https://arxiv.org/abs/2303.16118 发表或投稿:无 代码:https://github.com/MCG-NJU/CycleACR. [11] LinK: Linear Kernel for LiDAR-based 3D Perception 标题:LinK:基于激光雷达的3D感知的线性内核 链接:https://arxiv.org/abs/2303.16094 发表或投稿:CVPR 代码:https://github.com/MCG-NJU/LinK. [12] Rethinking matching-based few-shot action recognition 标题:基于匹配的少镜头动作识别的再思考 链接:https://arxiv.org/abs/2303.16084 发表或投稿:无 代码:https://jbertrand89.github.io/matching-based-fsar [13] Efficient solutions to the relative pose of three calibrated cameras from four points using virtual correspondences 标题:使用虚拟对应对来自四个点的三个校准相机的相对姿态的有效解决方案 链接:https://arxiv.org/abs/2303.16078 发表或投稿:无 代码:未开源 [14] Real-time Multi-person Eyeblink Detection in the Wild for Untrimmed Video 标题:用于无边框视频的野外实时多人眨眼检测 链接:https://arxiv.org/abs/2303.16053 发表或投稿:CVPR 代码:未开源 [15] F$^{2}$-NeRF: Fast Neural Radiance Field Training with Free Camera Trajectories 标题:F$^{2}$-NeRF:使用自由相机轨迹的快速神经辐射场训练 链接:https://arxiv.org/abs/2303.15951 发表或投稿:CVPR 代码:https://totoro97.github.io/projects/f2-nerf. [16] Deep Selection: A Fully Supervised Camera Selection Network for Surgery Recordings 标题:深度选择:一个完全监督的手术记录摄像机选择网络 链接:https://arxiv.org/abs/2303.15947 发表或投稿:无 代码:未开源 [17] VIVE3D: Viewpoint-Independent Video Editing using 3D-Aware GANs 标题:VIVE3D:使用3D感知GANs进行视点独立视频编辑 链接:https://arxiv.org/abs/2303.15893 发表或投稿:CVPR 代码:未开源 [18] Head3D: Complete 3D Head Generation via Tri-plane Feature Distillation 标题:Head3D:通过三平面特征提取完成3D头部生成 链接:https://arxiv.org/abs/2303.15892 发表或投稿:无 代码:未开源 [19] Novel View Synthesis of Humans using Differentiable Rendering 标题:使用可微分绘制的新型人类视图合成 链接:https://arxiv.org/abs/2303.15880 发表或投稿:无 代码:https://github.com/GuillaumeRochette/HumanViewSynthesis. [20] STMixer: A One-Stage Sparse Action Detector 标题:STMixer:一种单级稀疏动作检测器 链接:https://arxiv.org/abs/2303.15879 发表或投稿:CVPR 代码:未开源 [21] StarNet: Style-Aware 3D Point Cloud Generation 标题:StarNet:风格感知的3D点云生成 链接:https://arxiv.org/abs/2303.15805 发表或投稿:无 代码:未开源 [22] Instruct 3D-to-3D: Text Instruction Guided 3D-to-3D conversion 标题:指导3D到3D:文本指导引导的3D到3D转换 链接:https://arxiv.org/abs/2303.15780 发表或投稿:无 代码:https://sony.github.io/Instruct3Dto3D-doc/ [23] System-status-aware Adaptive Network for Online Streaming Video Understanding 标题:用于在线流媒体视频理解的系统状态感知自适应网络 链接:https://arxiv.org/abs/2303.15742 发表或投稿:CVPR 代码:未开源 [24] Cross-View Visual Geo-Localization for Outdoor Augmented Reality 标题:用于户外增强现实的跨视图视觉地理定位 链接:https://arxiv.org/abs/2303.15676 发表或投稿:无 代码:未开源 [25] Fine-grained Audible Video Description 标题:细粒度音频视频描述 链接:https://arxiv.org/abs/2303.15616 发表或投稿:CVPR 代码:未开源 [26] OmniAvatar: Geometry-Guided Controllable 3D Head Synthesis 标题:OmniAvatar:几何引导可控3D头部合成 链接:https://arxiv.org/abs/2303.15539 发表或投稿:无 代码:未开源 [27] Sat2Density: Faithful Density Learning from Satellite-Ground Image Pairs 标题:Sat2Density:从卫星地面图像对中进行忠实密度学习 链接:https://arxiv.org/abs/2303.14672 发表或投稿:无 代码:https://sat2density.github.io [28] Diverse Motion In-betweening with Dual Posture Stitching 标题:采用双体式缝合,实现多样化的中间运动 链接:https://arxiv.org/abs/2303.14457 发表或投稿:无 代码:未开源 [29] PACE: Data-Driven Virtual Agent Interaction in Dense and Cluttered Environments 标题:PACE:密集和混乱环境中数据驱动的虚拟代理交互 链接:https://arxiv.org/abs/2303.14255 发表或投稿:无 代码:https://gamma.umd.edu/pace/. [30] STB-VMM: Swin Transformer Based Video Motion Magnification 标题:STB-VMM:基于Swin Transformer的视频运动放大 链接:https://arxiv.org/abs/2302.10001 发表或投稿:无 代码:https://github.com/RLado/STB-VMM [31] HexPlane: A Fast Representation for Dynamic Scenes 标题:HexPlane:动态场景的快速表示 链接:https://arxiv.org/abs/2301.09632 发表或投稿:CVPR 代码:https://caoang327.github.io/HexPlane [32] Anchor3DLane: Learning to Regress 3D Anchors for Monocular 3D Lane Detection 标题:Anchor3DLane:学习回归3D锚用于单目3D车道检测 链接:https://arxiv.org/abs/2301.02371 发表或投稿:CVPR 代码:https://github.com/tusen-ai/Anchor3DLane. [33] Shakes on a Plane: Unsupervised Depth Estimation from Unstabilized Photography 标题:平面上的抖动:来自非稳定摄影的无监督深度估计 链接:https://arxiv.org/abs/2212.12324 发表或投稿:无 代码:https://light.princeton.edu/publication/soap [34] GazeNeRF: 3D-Aware Gaze Redirection with Neural Radiance Fields 标题:GazeNeRF:具有神经辐射场的3D感知凝视重定向 链接:https://arxiv.org/abs/2212.04823 发表或投稿:CVPR 代码:https://github.com/AlessandroRuzzi/GazeNeRF [35] EditableNeRF: Editing Topologically Varying Neural Radiance Fields by Key Points 标题:EditableNeRF:按关键点编辑拓扑变化的神经辐射场 链接:https://arxiv.org/abs/2212.04247 发表或投稿:CVPR 代码:https://chengwei-zheng.github.io/EditableNeRF/. [36] SliceMatch: Geometry-guided Aggregation for Cross-View Pose Estimation 标题:SliceMatch:用于交叉视图姿态估计的几何引导聚合 链接:https://arxiv.org/abs/2211.14651 发表或投稿:无 代码:未开源 [37] Re^2TAL: Rewiring Pretrained Video Backbones for Reversible Temporal Action Localization 标题:Re^2TAL:重新连接预训练的视频主干,实现可逆的时间动作定位 链接:https://arxiv.org/abs/2211.14053 发表或投稿:无 代码:未开源 [38] PermutoSDF: Fast Multi-View Reconstruction with Implicit Surfaces using Permutohedral Lattices 标题:PermutoSDF:使用Permut面体格的隐式曲面快速多视图重建 链接:https://arxiv.org/abs/2211.12562 发表或投稿:CVPR 代码:https://radualexandru.github.io/permuto_sdf [39] MegaPortraits: One-shot Megapixel Neural Head Avatars 标题:MegaPortraits:一次拍摄的百万像素神经头头像 链接:https://arxiv.org/abs/2207.07621 发表或投稿:无 代码:未开源 [40] Real-World Deep Local Motion Deblurring 标题:真实世界深层局部运动去模糊 链接:https://arxiv.org/abs/2204.08179 发表或投稿:AAAI 代码:未开源 [41] LASER: LAtent SpacE Rendering for 2D Visual Localization 标题:LASER:LAtent SpacE渲染用于2D视觉定位 链接:https://arxiv.org/abs/2204.00157 发表或投稿:CVPR 代码:未开源 [42] Multi-sensor large-scale dataset for multi-view 3D reconstruction 标题:用于多视图三维重建的多传感器大规模数据集 链接:https://arxiv.org/abs/2203.06111 发表或投稿:无 代码:未开源 image segmentation相关(12篇)[1] OpenInst: A Simple Query-Based Method for Open-World Instance Segmentation 标题:OpenInst:一种简单的基于查询的开放世界实例分割方法 链接:https://arxiv.org/abs/2303.15859 发表或投稿:无 代码:未开源 [2] FMAS: Fast Multi-Objective SuperNet Architecture Search for Semantic Segmentation 标题:FMAS:用于语义分割的快速多目标SuperNet体系结构搜索 链接:https://arxiv.org/abs/2303.16322 发表或投稿:无 代码:未开源 [3] Medical Image Analysis using Deep Relational Learning 标题:利用深度关系学习进行医学图像分析 链接:https://arxiv.org/abs/2303.16099 发表或投稿:无 代码:未开源 [4] Mask-Free Video Instance Segmentation 标题:无掩码视频实例分割 链接:https://arxiv.org/abs/2303.15904 发表或投稿:CVPR 代码:https://github.com/SysCV/MaskFreeVis. [5] That Label's Got Style: Handling Label Style Bias for Uncertain Image Segmentation 标题:标签的风格:处理不确定图像分割的标签风格偏差 链接:https://arxiv.org/abs/2303.15850 发表或投稿:无 代码:未开源 [6] AutoKary2022: A Large-Scale Densely Annotated Dateset for Chromosome Instance Segmentation 标题:AutoKary2022:一种用于染色体实例分割的大规模密集注释日期集 链接:https://arxiv.org/abs/2303.15839 发表或投稿:无 代码:https://github.com/wangjuncongyu/chromosome-instance-segmentation-dataset. [7] Deformable Kernel Expansion Model for Efficient Arbitrary-shaped Scene Text Detection 标题:用于高效任意形状场景文本检测的可变形内核扩展模型 链接:https://arxiv.org/abs/2303.15737 发表或投稿:无 代码:未开源 [8] Few-Shot 3D Point Cloud Semantic Segmentation via Stratified Class-Specific Attention Based Transformer Network 标题:基于分层类特定注意的Transformer网络的少镜头三维点云语义分割 链接:https://arxiv.org/abs/2303.15654 发表或投稿:无 代码:未开源 [9] 4D Panoptic Segmentation as Invariant and Equivariant Field Prediction 标题:作为不变和等变场预测的4D泛光分割 链接:https://arxiv.org/abs/2303.15651 发表或投稿:无 代码:未开源 [10] PolyFormer: Referring Image Segmentation as Sequential Polygon Generation 标题:PolyFormer:将图像分割称为顺序多边形生成 链接:https://arxiv.org/abs/2302.07387 发表或投稿:CVPR 代码:https://polyformer.github.io/ [11] Automatically Score Tissue Images Like a Pathologist by Transfer Learning 标题:通过迁移学习像病理学家一样自动对组织图像进行评分 链接:https://arxiv.org/abs/2209.05954 发表或投稿:无 代码:未开源 [12] Dynamic Focus-aware Positional Queries for Semantic Segmentation 标题:用于语义分割的动态焦点感知位置查询 链接:https://arxiv.org/abs/2204.01244 发表或投稿:CVPR 代码:https://github.com/ziplab/FASeg Diffusion相关(10篇)[1] Guided Depth Super-Resolution by Deep Anisotropic Diffusion 标题:深度各向异性扩散的制导深度超分辨率 链接:https://arxiv.org/abs/2211.11592 发表或投稿:CVPR 代码:https://github.com/prs-eth/Diffusion-Super-Resolution) [2] A Pilot Study of Query-Free Adversarial Attack against Stable Diffusion 标题:针对稳定扩散的无查询对抗攻击的初步研究 链接:https://arxiv.org/abs/2303.16378 发表或投稿:无 代码:未开源 [3] Flow supervision for Deformable NeRF 标题:可变形NeRF的流量监控 链接:https://arxiv.org/abs/2303.16333 发表或投稿:无 代码:未开源 [4] Visual Chain-of-Thought Diffusion Models 标题:视觉思维链扩散模型 链接:https://arxiv.org/abs/2303.16187 发表或投稿:无 代码:未开源 [5] DDMM-Synth: A Denoising Diffusion Model for Cross-modal Medical Image Synthesis with Sparse-view Measurement Embedding 标题:DDMM-Synth:一种用于稀疏视图测量嵌入的跨模态医学图像合成的去噪扩散模型 链接:https://arxiv.org/abs/2303.15770 发表或投稿:无 代码:未开源 [6] Modiff: Action-Conditioned 3D Motion Generation with Denoising Diffusion Probabilistic Models 标题:Modiff:具有去噪扩散概率模型的动作条件三维运动生成 链接:https://arxiv.org/abs/2301.03949 发表或投稿:无 代码:未开源 [7] Fake it till you make it: Learning transferable representations from synthetic ImageNet clones 标题:伪造直到成功:从合成ImageNet克隆中学习可转移表示 链接:https://arxiv.org/abs/2212.08420 发表或投稿:CVPR 代码:https://europe.naverlabs.com/imagenet-sd/ [8] DiffPose: Toward More Reliable 3D Pose Estimation 标题:DiffPose:实现更可靠的3D姿态估计 链接:https://arxiv.org/abs/2211.16940 发表或投稿:CVPR 代码:https://gongjia0208.github.io/Diffpose/. [9] CRAFT: Concept Recursive Activation FacTorization for Explainability 标题:CRAFT:可解释性的概念递归激活法 链接:https://arxiv.org/abs/2211.10154 发表或投稿:Pattern Recognition 代码:未开源 [10] Seeing Beyond the Brain: Conditional Diffusion Model with Sparse Masked Modeling for Vision Decoding 标题:超越大脑的视觉:用于视觉解码的稀疏掩模条件扩散模型 链接:https://arxiv.org/abs/2211.06956 发表或投稿:CVPR 代码:https://mind-vis.github.io/ Others相关(43篇)[1] Two-Stage Context-Aware model for Predicting Future Motion of Dynamic Agents 标题:预测动态代理未来运动的两阶段上下文感知模型 链接:https://arxiv.org/abs/2211.08609 发表或投稿:无 代码:未开源 [2] Problems and shortcuts in deep learning for screening mammography 标题:乳腺钼靶筛查深度学习中的问题和捷径 链接:https://arxiv.org/abs/2303.16417 发表或投稿:无 代码:未开源 [3] Crime Prediction Using Machine Learning and Deep Learning: A Systematic Review and Future Directions 标题:利用机器学习和深度学习进行犯罪预测:系统综述和未来方向 链接:https://arxiv.org/abs/2303.16310 发表或投稿:无 代码:未开源 [4] Data Efficient Contrastive Learning in Histopatholgy using Active Sampling 标题:使用主动抽样的组织病理学中的数据高效对比学习 链接:https://arxiv.org/abs/2303.16247 发表或投稿:无 代码:未开源 [5] CuNeRF: Cube-Based Neural Radiance Field for Zero-Shot Medical Image Arbitrary-Scale Super Resolution 标题:CuNeRF:基于立方的零样本医学图像任意尺度超分辨率神经辐射场 链接:https://arxiv.org/abs/2303.16242 发表或投稿:无 代码:未开源 [6] Spatiotemporal Self-supervised Learning for Point Clouds in the Wild 标题:野外点云的时空自监督学习 链接:https://arxiv.org/abs/2303.16235 发表或投稿:CVPR 代码:未开源 [7] CCuantuMM: Cycle-Consistent Quantum-Hybrid Matching of Multiple Shapes 标题:CCuantuMM:多个形状的循环一致量子混合匹配 链接:https://arxiv.org/abs/2303.16202 发表或投稿:Pattern Recognition 代码:https://4dqv.mpi-inf.mpg.de/CCuantuMM/ [8] Evaluating the Effectiveness of 2D and 3D Features for Predicting Tumor Response to Chemotherapy 标题:评估2D和3D特征预测肿瘤对化疗反应的有效性 链接:https://arxiv.org/abs/2303.16123 发表或投稿:无 代码:未开源 [9] Neural Collapse Inspired Federated Learning with Non-iid Data 标题:基于神经崩溃的非iid数据联合学习 链接:https://arxiv.org/abs/2303.16066 发表或投稿:无 代码:未开源 [10] HiLo: Exploiting High Low Frequency Relations for Unbiased Panoptic Scene Graph Generation 标题:HiLo:利用高低频关系生成无偏全景场景图 链接:https://arxiv.org/abs/2303.15994 发表或投稿:无 代码:未开源 [11] UFO: A unified method for controlling Understandability and Faithfulness Objectives in concept-based explanations for CNNs 标题:不明飞行物:一种统一的方法,用于控制细胞神经网络基于概念的解释中的可理解性和真实性目标 链接:https://arxiv.org/abs/2303.15632 发表或投稿:无 代码:未开源 [12] TOFA: Transfer-Once-for-All 标题:TOFA:一次性转移 链接:https://arxiv.org/abs/2303.15485 发表或投稿:无 代码:未开源 [13] Learning Rotation-Equivariant Features for Visual Correspondence 标题:学习视觉对应的旋转等变特征 链接:https://arxiv.org/abs/2303.15472 发表或投稿:CVPR 代码:未开源 [14] Enlarging Instance-specific and Class-specific Information for Open-set Action Recognition 标题:扩展实例特定信息和类特定信息用于开放集动作识别 链接:https://arxiv.org/abs/2303.15467 发表或投稿:CVPR 代码:https://github.com/Jun-CEN/PSL. [15] Comparison between layer-to-layer network training and conventional network training using Convolutional Neural Networks 标题:使用卷积神经网络进行分层网络训练与传统网络训练的比较 链接:https://arxiv.org/abs/2303.15245 发表或投稿:无 代码:未开源 [16] One-shot Feature-Preserving Point Cloud Simplification with Gaussian Processes on Riemannian Manifolds 标题:黎曼流形上高斯过程的一次特征保持点云简化 链接:https://arxiv.org/abs/2303.15225 发表或投稿:无 代码:未开源 [17] Automatic breach detection during spine pedicle drilling based on vibroacoustic sensing 标题:基于振动声传感的椎弓根钻孔过程中的自动缺口检测 链接:https://arxiv.org/abs/2303.15114 发表或投稿:无 代码:未开源 [18] Global Relation Modeling and Refinement for Bottom-Up Human Pose Estimation 标题:自下而上的人体姿态估计的全局关系建模与优化 链接:https://arxiv.org/abs/2303.14888 发表或投稿:无 代码:未开源 [19] The Resource Problem of Using Linear Layer Leakage Attack in Federated Learning 标题:联邦学习中使用线性层泄漏攻击的资源问题 链接:https://arxiv.org/abs/2303.14868 发表或投稿:CVPR 代码:未开源 [20] Multi-Phase Relaxation Labeling for Square Jigsaw Puzzle Solving 标题:求解方形拼图的多相松弛标记 链接:https://arxiv.org/abs/2303.14793 发表或投稿:无 代码:未开源 [21] Analyzing Effects of Mixed Sample Data Augmentation on Model Interpretability 标题:混合样本数据扩充对模型可解释性的影响分析 链接:https://arxiv.org/abs/2303.14608 发表或投稿:无 代码:未开源 [22] SIO: Synthetic In-Distribution Data Benefits Out-of-Distribution Detection 标题:SIO:合成分布内数据优势分布外检测 链接:https://arxiv.org/abs/2303.14531 发表或投稿:无 代码:https://github.com/zjysteven/SIO. [23] Deep Active Learning with Contrastive Learning Under Realistic Data Pool Assumptions 标题:现实数据池假设下的深度主动学习与对比学习 链接:https://arxiv.org/abs/2303.14433 发表或投稿:AAAI 代码:未开源 [24] On Function-Coupled Watermarks for Deep Neural Networks 标题:关于深度神经网络的函数耦合水印 链接:https://arxiv.org/abs/2302.10296 发表或投稿:无 代码:https://github.com/cure-lab/Function-Coupled-Watermark. [25] Stitchable Neural Networks 标题:可缝合神经网络 链接:https://arxiv.org/abs/2302.06586 发表或投稿:CVPR 代码:https://snnet.github.io/ [26] High-fidelity Interpretable Inverse Rig: An Accurate and Sparse Solution Optimizing the Quartic Blendshape Model 标题:高保真度可解释反演装置:优化四次混合形状模型的精确稀疏解 链接:https://arxiv.org/abs/2302.04820 发表或投稿:无 代码:未开源 [27] Methods and Tools for Monitoring Driver's Behavior 标题:监控驾驶员行为的方法和工具 链接:https://arxiv.org/abs/2301.12269 发表或投稿:无 代码:未开源 [28] From Plate to Prevention: A Dietary Nutrient-aided Platform for Health Promotion in Singapore 标题:从盘子到预防:新加坡健康促进的膳食营养辅助平台 链接:https://arxiv.org/abs/2301.03829 发表或投稿:无 代码:https://foodlg.comp.nus.edu.sg/. [29] ARO-Net: Learning Implicit Fields from Anchored Radial Observations 标题:ARO-Net:从锚定径向观测中学习隐式场 链接:https://arxiv.org/abs/2212.10275 发表或投稿:CVPR 代码:https://github.com/yizhiwang96/ARO-Net [30] RepMode: Learning to Re-parameterize Diverse Experts for Subcellular Structure Prediction 标题:RepMode:学习重新参数化用于亚细胞结构预测的不同专家 链接:https://arxiv.org/abs/2212.10066 发表或投稿:CVPR 代码:未开源 [31] The Differentiable Lens: Compound Lens Search over Glass Surfaces and Materials for Object Detection 标题:可微分透镜:用于物体检测的玻璃表面和材料上的复合透镜搜索 链接:https://arxiv.org/abs/2212.04441 发表或投稿:CVPR 代码:未开源 [32] Generalizing Gaze Estimation with Weak-Supervision from Synthetic Views 标题:从综合角度推广弱监督注视估计 链接:https://arxiv.org/abs/2212.02997 发表或投稿:无 代码:未开源 [33] Exact-NeRF: An Exploration of a Precise Volumetric Parameterization for Neural Radiance Fields 标题:精确NeRF:神经辐射场精确体积参数化的探索 链接:https://arxiv.org/abs/2211.12285 发表或投稿:无 代码:未开源 [34] PLIKS: A Pseudo-Linear Inverse Kinematic Solver for 3D Human Body Estimation 标题:PLIKS:一种用于三维人体估计的伪线性逆运动学解算器 链接:https://arxiv.org/abs/2211.11734 发表或投稿:CVPR 代码:未开源 [35] DeS3: Attention-driven Self and Soft Shadow Removal using ViT Similarity and Color Convergence 标题:DeS3:使用ViT相似性和颜色收敛的注意力驱动的自阴影和软阴影去除 链接:https://arxiv.org/abs/2211.08089 发表或投稿:无 代码:未开源 [36] LPT: Long-tailed Prompt Tuning for Image Classification 标题:LPT:图像分类的长尾提示调整 链接:https://arxiv.org/abs/2210.01033 发表或投稿:ICLR 代码:未开源 [37] Bias Mimicking: A Simple Sampling Approach for Bias Mitigation 标题:偏差模拟:一种用于减少偏差的简单采样方法 链接:https://arxiv.org/abs/2209.15605 发表或投稿:无 代码:https://github.com/mqraitem/Bias-Mimicking} [38] DETRs with Hybrid Matching 标题:具有混合匹配的DETR 链接:https://arxiv.org/abs/2207.13080 发表或投稿:CVPR 代码:https://github.com/HDETR [39] Learning Robust Representation for Joint Grading of Ophthalmic Diseases via Adaptive Curriculum and Feature Disentanglement 标题:基于自适应课程和特征纠缠的眼科疾病联合评分的学习鲁棒表示 链接:https://arxiv.org/abs/2207.04183 发表或投稿:无 代码:未开源 [40] MobileOne: An Improved One millisecond Mobile Backbone 标题:MobileOne:一种改进的1毫秒移动主干 链接:https://arxiv.org/abs/2206.04040 发表或投稿:CVPR 代码:https://github.com/apple/ml-mobileone [41] PointVector: A Vector Representation In Point Cloud Analysis 标题:点向量:点云分析中的一种向量表示 链接:https://arxiv.org/abs/2205.10528 发表或投稿:CVPR 代码:未开源 [42] Cerebral Palsy Prediction with Frequency Attention Informed Graph Convolutional Networks 标题:频率-注意力知情图卷积网络预测脑瘫 链接:https://arxiv.org/abs/2204.10997 发表或投稿:无 代码:未开源 [43] Iterative label cleaning for transductive and semi-supervised few-shot learning 标题:用于转导和半监督少镜头学习的迭代标签清洗 链接:https://arxiv.org/abs/2012.07962 发表或投稿:ICCV 代码:https://github.com/MichalisLazarou/iLPC. Open Domain|Domain Adaptation|Open Vocalbulary相关(7篇)[1] Global Adaptation meets Local Generalization: Unsupervised Domain Adaptation for 3D Human Pose Estimation 标题:全局自适应与局部泛化:用于三维人体姿态估计的无监督域自适应 链接:https://arxiv.org/abs/2303.16456 发表或投稿:无 代码:未开源 [2] Domain Adaptive Semantic Segmentation by Optimal Transport 标题:基于最优传输的领域自适应语义分割 链接:https://arxiv.org/abs/2303.16435 发表或投稿:无 代码:未开源 [3] SFHarmony: Source Free Domain Adaptation for Distributed Neuroimaging Analysis 标题:SFHarmony:用于分布式神经成像分析的无源域自适应 链接:https://arxiv.org/abs/2303.15965 发表或投稿:无 代码:https://github.com/nkdinsdale/SFHarmony}. [4] Complementary Domain Adaptation and Generalization for Unsupervised Continual Domain Shift Learning 标题:无监督连续领域移位学习的互补领域自适应与泛化 链接:https://arxiv.org/abs/2303.15833 发表或投稿:无 代码:未开源 [5] MS-MT: Multi-Scale Mean Teacher with Contrastive Unpaired Translation for Cross-Modality Vestibular Schwannoma and Cochlea Segmentation 标题:MS-MT:跨模态前庭神经鞘瘤和耳蜗分割的多尺度均值教师对比非配对翻译 链接:https://arxiv.org/abs/2303.15826 发表或投稿:无 代码:未开源 [6] Few-Shot Domain Adaptation for Low Light RAW Image Enhancement 标题:用于微光RAW图像增强的少镜头域自适应 链接:https://arxiv.org/abs/2303.15528 发表或投稿:BMVC 代码:https://val.cds.iisc.ac.in/HDR/BMVC21/index.html [7] Domain-General Crowd Counting in Unseen Scenarios 标题:未知场景下的领域通用人群计数 链接:https://arxiv.org/abs/2212.02573 发表或投稿:AAAI 代码:未开源 Transformer相关(13篇)[1] Self-positioning Point-based Transformer for Point Cloud Understanding 标题:用于点云理解的自定位基于点的转换器 链接:https://arxiv.org/abs/2303.16450 发表或投稿:CVPR 代码:https://github.com/mlvlab/SPoTr. [2] SnakeVoxFormer: Transformer-based Single Image\\Voxel Reconstruction with Run Length Encoding 标题:SnakeVoxFormer:基于变换器的游程编码单图像\\体素重建 链接:https://arxiv.org/abs/2303.16293 发表或投稿:无 代码:未开源 [3] CryoFormer: Continuous Reconstruction of 3D Structures from Cryo-EM Data using Transformer-based Neural Representations 标题:CryoFormer:使用基于变压器的神经表示从Cryo-EM数据连续重建3D结构 链接:https://arxiv.org/abs/2303.16254 发表或投稿:无 代码:https://cryoformer.github.io. [4] Transferable Adversarial Attacks on Vision Transformers with Token Gradient Regularization 标题:基于令牌梯度正则化的视觉变换器的可转移对抗攻击 链接:https://arxiv.org/abs/2303.15754 发表或投稿:CVPR 代码:未开源 [5] MeMaHand: Exploiting Mesh-Mano Interaction for Single Image Two-Hand Reconstruction 标题:MeMaHand:利用Mesh Mano交互进行单图像双手重建 链接:https://arxiv.org/abs/2303.15718 发表或投稿:无 代码:未开源 [6] TFS-ViT: Token-Level Feature Stylization for Domain Generalization 标题:TFS ViT:用于领域泛化的令牌级特征样式化 链接:https://arxiv.org/abs/2303.15698 发表或投稿:无 代码:https://github.com/Mehrdad-Noori/TFS-ViT_Token-level_Feature_Stylization. [7] Learning Expressive Prompting With Residuals for Vision Transformers 标题:学习视觉变换器的残差表达提示 链接:https://arxiv.org/abs/2303.15591 发表或投稿:CVPR 代码:未开源 [8] D-TrAttUnet: Dual-Decoder Transformer-Based Attention Unet Architecture for Binary and Multi-classes Covid-19 Infection Segmentation 标题:D-TrAttUnet:基于双解码变换的二类和多类新冠肺炎感染分割的注意Unet结构 链接:https://arxiv.org/abs/2303.15576 发表或投稿:无 代码:未开源 [9] MoViT: Memorizing Vision Transformers for Medical Image Analysis 标题:MoViT:用于医学图像分析的记忆视觉转换器 链接:https://arxiv.org/abs/2303.15553 发表或投稿:无 代码:未开源 [10] Semantic-visual Guided Transformer for Few-shot Class-incremental Learning 标题:用于少镜头类增量学习的语义视觉引导转换器 链接:https://arxiv.org/abs/2303.15494 发表或投稿:无 代码:未开源 [11] Supervised Masked Knowledge Distillation for Few-Shot Transformers 标题:少镜头变形金刚的监督蒙面知识提取 链接:https://arxiv.org/abs/2303.15466 发表或投稿:CVPR 代码:https://github.com/HL-hanlin/SMKD. [12] Connecting the Dots: Floorplan Reconstruction Using Two-Level Queries 标题:连接点:使用两级查询重建平面图 链接:https://arxiv.org/abs/2211.15658 发表或投稿:CVPR 代码:https://github.com/ywyue/RoomFormer. [13] Deep Convolutional Pooling Transformer for Deepfake Detection 标题:用于深度伪造检测的深度卷积池变换器 链接:https://arxiv.org/abs/2209.05299 发表或投稿:无 代码:未开源 Image Reconstruction|Image Denoising|Image Compression相关(33篇)[1] Random Weights Networks Work as Loss Prior Constraint for Image Restoration 标题:随机权重网络作为图像恢复的损失先验约束 链接:https://arxiv.org/abs/2303.16438 发表或投稿:无 代码:未开源 [2] Real-time Controllable Denoising for Image and Video 标题:图像和视频的实时可控去噪 链接:https://arxiv.org/abs/2303.16425 发表或投稿:CVPR 代码:未开源 [3] Learning Iterative Neural Optimizers for Image Steganography 标题:用于图像隐写术的学习迭代神经优化算法 链接:https://arxiv.org/abs/2303.16206 发表或投稿:ICLR 代码:未开源 [4] Whole-body PET image denoising for reduced acquisition time 标题:减少采集时间的全身PET图像去噪 链接:https://arxiv.org/abs/2303.16085 发表或投稿:无 代码:未开源 [5] fRegGAN with K-space Loss Regularization for Medical Image Translation 标题:用于医学图像翻译的具有K空间损失正则化的fRegGAN 链接:https://arxiv.org/abs/2303.15938 发表或投稿:无 代码:未开源 [6] Hyperbolic Geometry in Computer Vision: A Novel Framework for Convolutional Neural Networks 标题:计算机视觉中的双曲几何:卷积神经网络的一种新框架 链接:https://arxiv.org/abs/2303.15919 发表或投稿:无 代码:https://github.com/kschwethelm/HyperbolicCV. [7] 4K-HAZE: A Dehazing Benchmark with 4K Resolution Hazy and Haze-Free Images 标题:4K-HAZE:4K分辨率模糊和无雾图像的去雾基准 链接:https://arxiv.org/abs/2303.15848 发表或投稿:无 代码:未开源 [8] Make the Most Out of Your Net: Alternating Between Canonical and Hard Datasets for Improved Image Demosaicing 标题:充分利用你的网络:在标准数据集和硬数据集之间交替,以改进图像去马赛克 链接:https://arxiv.org/abs/2303.15792 发表或投稿:无 代码:未开源 [9] SVD-DIP: Overcoming the Overfitting Problem in DIP-based CT Reconstruction 标题:SVD-DIP:克服基于DIP的CT重建中的过度拟合问题 链接:https://arxiv.org/abs/2303.15748 发表或投稿:无 代码:未开源 [10] Explicit Attention-Enhanced Fusion for RGB-Thermal Perception Tasks 标题:RGB热感任务的显式注意力增强融合 链接:https://arxiv.org/abs/2303.15710 发表或投稿:无 代码:https://github.com/FreeformRobotics/EAEFNet. [11] Mask and Restore: Blind Backdoor Defense at Test Time with Masked Autoencoder 标题:掩蔽和恢复:在测试时使用掩蔽自动编码器进行盲后门防御 链接:https://arxiv.org/abs/2303.15564 发表或投稿:无 代码:https://github.com/tsun/BDMAE. [12] Quantum Multi-Model Fitting 标题:量子多模型拟合 链接:https://arxiv.org/abs/2303.15444 发表或投稿:Pattern Recognition 代码:https://github.com/FarinaMatteo/qmmf. [13] Human Pose Estimation in Extremely Low-Light Conditions 标题:极低光照条件下的人体姿态估计 链接:https://arxiv.org/abs/2303.15410 发表或投稿:CVPR 代码:未开源 [14] Recognizing Rigid Patterns of Unlabeled Point Clouds by Complete and Continuous Isometry Invariants with no False Negatives and no False Positives 标题:用不含假阴性和假阳性的完全连续等距不变量识别无标记点云的刚性模式 链接:https://arxiv.org/abs/2303.15385 发表或投稿:Pattern Recognition 代码:https://cvpr2023.thecvf.com. [15] DANI-Net: Uncalibrated Photometric Stereo by Differentiable Shadow Handling, Anisotropic Reflectance Modeling, and Neural Inverse Rendering 标题:DANI Net:通过可微分阴影处理、各向异性反射建模和神经逆向渲染实现的未校准光度立体 链接:https://arxiv.org/abs/2303.15101 发表或投稿:CVPR 代码:未开源 [16] Intersection over Union with smoothing for bounding box regression 标题:带边界框回归平滑的并集上的交集 链接:https://arxiv.org/abs/2303.15067 发表或投稿:无 代码:未开源 [17] CRRS: Concentric Rectangles Regression Strategy for Multi-point Representation on Fisheye Images 标题:CRRS:用于鱼眼图像多点表示的同心矩形回归策略 链接:https://arxiv.org/abs/2303.14639 发表或投稿:无 代码:未开源 [18] VisCo Grids: Surface Reconstruction with Viscosity and Coarea Grids 标题:VisCo网格:使用粘度和面积网格进行表面重建 链接:https://arxiv.org/abs/2303.14569 发表或投稿:NeurIPS 代码:未开源 [19] IDGI: A Framework to Eliminate Explanation Noise from Integrated Gradients 标题:IDGI:一个消除综合梯度解释噪声的框架 链接:https://arxiv.org/abs/2303.14242 发表或投稿:CVPR 代码:未开源 [20] Curricular Contrastive Regularization for Physics-aware Single Image Dehazing 标题:物理感知单图像去雾的课程对比正则化 链接:https://arxiv.org/abs/2303.14218 发表或投稿:CVPR 代码:未开源 [21] SU-Net: Pose estimation network for non-cooperative spacecraft on-orbit 标题:SU-Net:轨道上非合作航天器的姿态估计网络 链接:https://arxiv.org/abs/2302.10602 发表或投稿:无 代码:未开源 [22] MN-Pair Contrastive Damage Representation and Clustering for Prognostic Explanation 标题:MN对对比损伤表示和聚类用于预测解释 链接:https://arxiv.org/abs/2301.06077 发表或投稿:无 代码:未开源 [23] Efficient On-device Training via Gradient Filtering 标题:通过梯度滤波进行高效的设备上训练 链接:https://arxiv.org/abs/2301.00330 发表或投稿:CVPR 代码:未开源 [24] Improving Visual Representation Learning through Perceptual Understanding 标题:通过感知理解提高视觉表征学习 链接:https://arxiv.org/abs/2212.14504 发表或投稿:CVPR 代码:未开源 [25] Plateau-reduced Differentiable Path Tracing 标题:高原减少的可微分路径跟踪 链接:https://arxiv.org/abs/2211.17263 发表或投稿:CVPR 代码:https://mfischer-ucl.github.io/prdpt/ [26] Deep Curvilinear Editing: Commutative and Nonlinear Image Manipulation for Pretrained Deep Generative Model 标题:深度曲线编辑:预训练深度生成模型的交换和非线性图像处理 链接:https://arxiv.org/abs/2211.14573 发表或投稿:无 代码:未开源 [27] BiasBed -- Rigorous Texture Bias Evaluation 标题:BiasBed——严格的纹理偏移评估 链接:https://arxiv.org/abs/2211.13190 发表或投稿:无 代码:https://github.com/D1noFuzi/BiasBed [28] MIMT: Multi-Illuminant Color Constancy via Multi-Task Learning 标题:MIMT:通过多任务学习实现多光源颜色恒定性 链接:https://arxiv.org/abs/2211.08772 发表或投稿:无 代码:未开源 [29] DSCA: A Dual-Stream Network with Cross-Attention on Whole-Slide Image Pyramids for Cancer Prognosis 标题:DSCA:一个交叉关注全滑动图像金字塔的双流网络,用于癌症预测 链接:https://arxiv.org/abs/2206.05782 发表或投稿:无 代码:未开源 [30] Unsupervised Representation Learning for 3D MRI Super Resolution with Degradation Adaptation 标题:具有退化自适应的3D MRI超分辨率无监督表示学习 链接:https://arxiv.org/abs/2205.06891 发表或投稿:无 代码:未开源 [31] A Note on the Regularity of Images Generated by Convolutional Neural Networks 标题:关于卷积神经网络生成图像的正则性的一点注记 链接:https://arxiv.org/abs/2204.10588 发表或投稿:无 代码:未开源 [32] Edge Guided GANs with Contrastive Learning for Semantic Image Synthesis 标题:用于语义图像合成的具有对比学习的边缘引导GANs 链接:https://arxiv.org/abs/2003.13898 发表或投稿:无 代码:未开源 [33] A Grid-based Method for Removing Overlaps of Dimensionality Reduction Scatterplot Layouts 标题:一种基于网格的降维散点图布局重叠消除方法 链接:https://arxiv.org/abs/1903.06262 发表或投稿:无 代码:未开源 Image Retrieval|ReID|Face Recognition相关(7篇)[1] Facial recognition technology can expose political orientation from facial images even when controlling for demographics and self-presentation 标题:即使在控制人口统计和自我展示的情况下,面部识别技术也可以从面部图像中暴露政治取向 链接:https://arxiv.org/abs/2303.16343 发表或投稿:无 代码:未开源 [2] Large-scale Training Data Search for Object Re-identification 标题:用于对象重新识别的大规模训练数据搜索 链接:https://arxiv.org/abs/2303.16186 发表或投稿:CVPR 代码:https://github.com/yorkeyao/SnP. [3] RobustSwap: A Simple yet Robust Face Swapping Model against Attribute Leakage 标题:RobustSwap:一种简单而稳健的属性泄漏人脸交换模型 链接:https://arxiv.org/abs/2303.15768 发表或投稿:无 代码:https://robustswap.github.io/ [4] Colo-SCRL: Self-Supervised Contrastive Representation Learning for Colonoscopic Video Retrieval 标题:Colo SCRL:用于结肠镜视频检索的自监督对比表示学习 链接:https://arxiv.org/abs/2303.15671 发表或投稿:无 代码:未开源 [5] OTAvatar: One-shot Talking Face Avatar with Controllable Tri-plane Rendering 标题:OTAvatar:可控制三平面渲染的单镜头会说话的人脸头像 链接:https://arxiv.org/abs/2303.14662 发表或投稿:CVPR 代码:https://github.com/theEricMa/OTAvatar [6] Diverse Embedding Expansion Network and Low-Light Cross-Modality Benchmark for Visible-Infrared Person Re-identification 标题:用于可见红外人员重新识别的多样化嵌入扩展网络和微光交叉模态基准 链接:https://arxiv.org/abs/2303.14481 发表或投稿:CVPR 代码:https://github.com/ZYK100/LLCM [7] BaRe-ESA: A Riemannian Framework for Unregistered Human Body Shapes 标题:BaRe-ESA:未注册人体形状的黎曼框架 链接:https://arxiv.org/abs/2211.13185 发表或投稿:无 代码:未开源 Optical Character Recognition相关(1篇)[1] SynthRAD2023 Grand Challenge dataset: generating synthetic CT for radiotherapy 标题:SynthRAD2023 Grand Challenge数据集:生成用于放射治疗的合成CT 链接:https://arxiv.org/abs/2303.16320 发表或投稿:无 代码:https://doi.org/10.5281/zenodo.7260705) Knowledge|Distillation|Pruning|Graph相关(11篇)[1] Dice Semimetric Losses: Optimizing the Dice Score with Soft Labels 标题:骰子半度量损失:使用软标签优化骰子分数 链接:https://arxiv.org/abs/2303.16296 发表或投稿:无 代码:https://github.com/zifuwanggg/JDTLosses}{https://github.com/zifuwanggg/JDTLosses}. [2] Projected Latent Distillation for Data-Agnostic Consolidation in Distributed Continual Learning 标题:分布式连续学习中数据不可知整合的投影潜在蒸馏 链接:https://arxiv.org/abs/2303.15888 发表或投稿:无 代码:未开源 [3] Enhancing Depth Completion with Multi-View Monitored Distillation 标题:利用多视图监控蒸馏提高深度完井 链接:https://arxiv.org/abs/2303.15840 发表或投稿:无 代码:未开源 [4] Learning Second-Order Attentive Context for Efficient Correspondence Pruning 标题:学习二阶注意上下文进行有效的对应修剪 链接:https://arxiv.org/abs/2303.15761 发表或投稿:AAAI 代码:未开源 [5] DisWOT: Student Architecture Search for Distillation WithOut Training 标题:DisWOT:学生体系结构搜索,无需培训 链接:https://arxiv.org/abs/2303.15678 发表或投稿:CVPR 代码:https://lilujunai.github.io/DisWOT-CVPR2023/. [6] Sequential training of GANs against GAN-classifiers reveals correlated "knowledge gaps" present among independently trained GAN instances 标题:针对GAN分类器的GAN序列训练揭示了独立训练的GAN实例之间存在的相关“知识差距” 链接:https://arxiv.org/abs/2303.15533 发表或投稿:无 代码:未开源 [7] Binarizing Sparse Convolutional Networks for Efficient Point Cloud Analysis 标题:用于高效点云分析的稀疏卷积网络二元化 链接:https://arxiv.org/abs/2303.15493 发表或投稿:CVPR 代码:未开源 [8] Does `Deep Learning on a Data Diet' reproduce? Overall yes, but GraNd at Initialization does not 标题:“数据饮食上的深度学习”会重现吗?总体上是的,但初始化时的GraNd没有 链接:https://arxiv.org/abs/2303.14753 发表或投稿:无 代码:https://github.com/google/flax/commit/28fbd95500f4bf2f9924d2560062fa50e919b1a5). [9] Compacting Binary Neural Networks by Sparse Kernel Selection 标题:稀疏核选择压缩二元神经网络 链接:https://arxiv.org/abs/2303.14470 发表或投稿:CVPR 代码:未开源 [10] Jaccard Metric Losses: Optimizing the Jaccard Index with Soft Labels 标题:Jaccard度量损失:使用软标签优化Jaccard指数 链接:https://arxiv.org/abs/2302.05666 发表或投稿:ICML 代码:https://github.com/zifuwanggg/JDTLosses}{https://github.com/zifuwanggg/JDTLosses}. [11] Gabor Convolutional Networks 标题:Gabor卷积网络 链接:https://arxiv.org/abs/1705.01450 发表或投稿:无 代码:未开源 object detection相关(23篇)[1] ASIC: Aligning Sparse in-the-wild Image Collections 标题:ASIC:对齐野生图像集合中的稀疏 链接:https://arxiv.org/abs/2303.16201 发表或投稿:无 代码:https://kampta.github.io/asic}. [2] A Comparative Study of Federated Learning Models for COVID-19 Detection 标题:新冠肺炎检测联合学习模型的比较研究 链接:https://arxiv.org/abs/2303.16141 发表或投稿:无 代码:未开源 [3] GP3D: Generalized Pose Estimation in 3D Point Clouds: A case study on bin picking 标题:GP3D:三维点云中的广义姿态估计:一个关于bin picking的案例研究 链接:https://arxiv.org/abs/2303.16102 发表或投稿:无 代码:未开源 [4] HS-Pose: Hybrid Scope Feature Extraction for Category-level Object Pose Estimation 标题:HS姿态:用于类别级对象姿态估计的混合范围特征提取 链接:https://arxiv.org/abs/2303.15743 发表或投稿:Pattern Recognition 代码:未开源 [5] Object Discovery from Motion-Guided Tokens 标题:从运动引导令牌中发现对象 链接:https://arxiv.org/abs/2303.15555 发表或投稿:CVPR 代码:未开源 [6] CAMS: CAnonicalized Manipulation Spaces for Category-Level Functional Hand-Object Manipulation Synthesis 标题:CAMS:用于类别级功能手对象操作合成的CA非局部化操作空间 链接:https://arxiv.org/abs/2303.15469 发表或投稿:CVPR 代码:https://cams-hoi.github.io/ [7] Defect detection using weakly supervised learning 标题:使用弱监督学习的缺陷检测 链接:https://arxiv.org/abs/2303.15092 发表或投稿:无 代码:未开源 [8] A novel Multi to Single Module for small object detection 标题:一种用于小物体检测的新型多对单模块 链接:https://arxiv.org/abs/2303.14977 发表或投稿:无 代码:未开源 [9] Addressing the Challenges of Open-World Object Detection 标题:应对开放世界对象检测的挑战 链接:https://arxiv.org/abs/2303.14930 发表或投稿:无 代码:未开源 [10] Mind the Backbone: Minimizing Backbone Distortion for Robust Object Detection 标题:注意骨干:最大限度地减少骨干失真以实现稳健的目标检测 链接:https://arxiv.org/abs/2303.14744 发表或投稿:无 代码:未开源 [11] EfficientAD: Accurate Visual Anomaly Detection at Millisecond-Level Latencies 标题:EfficientAD:毫秒级延迟下的精确视觉异常检测 链接:https://arxiv.org/abs/2303.14535 发表或投稿:无 代码:未开源 [12] Waste Detection and Change Analysis based on Multispectral Satellite Imagery 标题:基于多光谱卫星图像的废弃物检测与变化分析 链接:https://arxiv.org/abs/2303.14521 发表或投稿:无 代码:未开源 [13] Visual-Tactile Sensing for In-Hand Object Reconstruction 标题:视觉触觉传感在手部物体重建中的应用 链接:https://arxiv.org/abs/2303.14498 发表或投稿:CVPR 代码:https://sites.google.com/view/vtaco/}. [14] Adaptive Sparse Convolutional Networks with Global Context Enhancement for Faster Object Detection on Drone Images 标题:具有全局上下文增强的自适应稀疏卷积网络用于无人机图像上的快速目标检测 链接:https://arxiv.org/abs/2303.14488 发表或投稿:CVPR 代码:https://github.com/Cuogeihong/CEASC. [15] Collaborative Multi-Object Tracking with Conformal Uncertainty Propagation 标题:具有保形不确定性传播的协同多目标跟踪 链接:https://arxiv.org/abs/2303.14346 发表或投稿:无 代码:未开源 [16] Learned Two-Plane Perspective Prior based Image Resampling for Efficient Object Detection 标题:基于学习双平面视角先验的图像重采样用于有效的目标检测 链接:https://arxiv.org/abs/2303.14311 发表或投稿:CVPR 代码:未开源 [17] Adaptive Base-class Suppression and Prior Guidance Network for One-Shot Object Detection 标题:用于一次射击目标检测的自适应基类抑制和先验制导网络 链接:https://arxiv.org/abs/2303.14240 发表或投稿:无 代码:未开源 [18] Breaking the "Object" in Video Object Segmentation 标题:打破视频对象分割中的“对象” 链接:https://arxiv.org/abs/2212.06200 发表或投稿:无 代码:未开源 [19] NeRF-RPN: A general framework for object detection in NeRFs 标题:NeRF RPN:NeRF中对象检测的通用框架 链接:https://arxiv.org/abs/2211.11646 发表或投稿:CVPR 代码:https://github.com/lyclyc52/NeRF_RPN. [20] Progressive Transformation Learning for Leveraging Virtual Images in Training 标题:在训练中利用虚拟图像的渐进式转换学习 链接:https://arxiv.org/abs/2211.01778 发表或投稿:CVPR 代码:未开源 [21] Consistent-Teacher: Towards Reducing Inconsistent Pseudo-targets in Semi-supervised Object Detection 标题:一致性教师:减少半监督目标检测中不一致的伪目标 链接:https://arxiv.org/abs/2209.01589 发表或投稿:CVPR 代码:https://github.com/Adamdad/ConsistentTeacher}. [22] SAFE: Sensitivity-Aware Features for Out-of-Distribution Object Detection 标题:SAFE:用于分布外对象检测的灵敏度感知功能 链接:https://arxiv.org/abs/2208.13930 发表或投稿:无 代码:未开源 [23] Neural Part Priors: Learning to Optimize Part-Based Object Completion in RGB-D Scans 标题:神经零件优先级:学习优化RGB-D扫描中基于零件的对象完成 链接:https://arxiv.org/abs/2203.09375 发表或投稿:CVPR 代码:未开源 Generative Adversarial Network相关(7篇)[1] Information-Theoretic GAN Compression with Variational Energy-based Model 标题:基于变分能量模型的信息论GAN压缩 链接:https://arxiv.org/abs/2303.16050 发表或投稿:Neurips 代码:未开源 [2] Towards Effective Adversarial Textured 3D Meshes on Physical Face Recognition 标题:面向物理人脸识别的有效对抗性纹理三维网格 链接:https://arxiv.org/abs/2303.15818 发表或投稿:CVPR 代码:未开源 [3] Improving the Transferability of Adversarial Samples by Path-Augmented Method 标题:用路径增强方法提高对抗样本的可转移性 链接:https://arxiv.org/abs/2303.15735 发表或投稿:CVPR 代码:未开源 [4] Learning the Unlearnable: Adversarial Augmentations Suppress Unlearnable Example Attacks 标题:学习不可学习:对抗性增强抑制不可学习的示例攻击 链接:https://arxiv.org/abs/2303.15127 发表或投稿:无 代码:https://github.com/lafeat/ueraser. [5] Improving the Transferability of Adversarial Examples via Direction Tuning 标题:通过方向调整提高对抗性例子的可迁移性 链接:https://arxiv.org/abs/2303.15109 发表或投稿:无 代码:未开源 [6] CFA: Class-wise Calibrated Fair Adversarial Training 标题:CFA:按级别校准的公平对抗性训练 链接:https://arxiv.org/abs/2303.14460 发表或投稿:CVPR 代码:https://github.com/PKU-ML/CFA}. [7] Utilizing Network Properties to Detect Erroneous Inputs 标题:利用网络财产检测错误输入 链接:https://arxiv.org/abs/2002.12520 发表或投稿:无 代码:未开源 Text Summarization相关(1篇)[1] SELF-VS: Self-supervised Encoding Learning For Video Summarization 标题:SELF-VS:视频摘要的自监督编码学习 链接:https://arxiv.org/abs/2303.15993 发表或投稿:无 代码:未开源 Image Classification|Image Recognition相关(7篇)[1] Exploring Deep Learning Methods for Classification of SAR Images: Towards NextGen Convolutions via Transformers 标题:探索SAR图像分类的深度学习方法:通过变压器实现下一代卷积 链接:https://arxiv.org/abs/2303.15852 发表或投稿:无 代码:未开源 [2] Automated wildlife image classification: An active learning tool for ecological applications 标题:自动野生动物图像分类:一种用于生态应用的主动学习工具 链接:https://arxiv.org/abs/2303.15823 发表或投稿:无 代码:未开源 [3] Iteratively Coupled Multiple Instance Learning from Instance to Bag Classifier for Whole Slide Image Classification 标题:用于全幻灯片图像分类的实例到袋分类器的迭代耦合多实例学习 链接:https://arxiv.org/abs/2303.15749 发表或投稿:无 代码:未开源 [4] Towards Artistic Image Aesthetics Assessment: a Large-scale Dataset and a New Method 标题:走向艺术形象美学评估:一个大规模数据集和一种新方法 链接:https://arxiv.org/abs/2303.15166 发表或投稿:CVPR 代码:https://github.com/Dreemurr-T/BAID.git [5] A Statistical Model for Predicting Generalization in Few-Shot Classification 标题:少镜头分类中预测泛化的统计模型 链接:https://arxiv.org/abs/2212.06461 发表或投稿:无 代码:未开源 [6] DynamicISP: Dynamically Controlled Image Signal Processor for Image Recognition 标题:动态ISP:用于图像识别的动态控制图像信号处理器 链接:https://arxiv.org/abs/2211.01146 发表或投稿:无 代码:未开源 [7] Pseudo-Data based Self-Supervised Federated Learning for Classification of Histopathological Images 标题:基于伪数据的自监督联合学习在组织病理学图像分类中的应用 链接:https://arxiv.org/abs/2205.15530 发表或投稿:无 代码:未开源 |
CopyRight 2018-2019 实验室设备网 版权所有 |