美文网首页
计算机视觉每日论文速递[10.02]

计算机视觉每日论文速递[10.02]

作者: arXiv每日论文速递 | 来源:发表于2019-10-02 10:38 被阅读0次

    同步公众号(arXiv每日论文速递),欢迎关注,感谢支持哦~

    [检测分类相关]:

    【1】 Research on insect pest image detection and recognition based on bio-inspired methods
    基于仿生方法的害虫图像检测与识别研究
    作者: Loris Nanni, Fabio Pancino
    链接:https://arxiv.org/abs/1910.00296

    【2】 Adversarial Patches Exploiting Contextual Reasoning in Object Detection
    目标检测中利用上下文推理的对抗性补丁
    作者: Aniruddha Saha, Hamed Pirsiavash
    链接:https://arxiv.org/abs/1910.00068

    [分割/语义相关]:

    【1】 Real-Time Semantic Stereo Matching
    实时语义立体匹配
    作者: Pier Luigi Dovesi, Stefano Mattoccia
    链接:https://arxiv.org/abs/1910.00541

    【2】 CapsuleVOS: Semi-Supervised Video Object Segmentation Using Capsule Routing
    CapsuleVOS:使用胶囊路由的半监督视频对象分割
    作者: Kevin Duarte, Mubarak Shah
    备注:8 pages, 6 figures, ICCV 2019
    链接:https://arxiv.org/abs/1910.00132

    【3】 LIP: Learning Instance Propagation for Video Object Segmentation
    LIP:用于视频对象分割的学习实例传播
    作者: Ye Lyu, Michael Ying Yang
    备注:ICCVW19
    链接:https://arxiv.org/abs/1910.00032

    【4】 Compensating Supervision Incompleteness with Prior Knowledge in Semantic Image Interpretation
    利用先验知识补偿语义图像解释中的监督不完全性
    作者: Ivan Donadello, Luciano Serafini
    链接:https://arxiv.org/abs/1910.00462

    [GAN/对抗式/生成式相关]:

    【1】 Deep Neural Rejection against Adversarial Examples
    针对对抗性实例的深层神经排斥
    作者: Angelo Sotgiu, Fabio Roli
    链接:https://arxiv.org/abs/1910.00470

    【2】 CMTS: Conditional Multiple Trajectory Synthesizer for Generating Safety-critical Driving Scenarios
    CMTS:用于生成安全关键驾驶场景的条件多轨迹合成器
    作者: Wenhao Ding, Ding Zhao
    备注:Submitted to ICRA 2020, 8 pages, 7 figures
    链接:https://arxiv.org/abs/1910.00099

    [半/弱/无监督相关]:

    【1】 Unsupervised Generative 3D Shape Learning from Natural Images
    基于自然图像的无监督生成三维形状学习
    作者: Attila Szabó, Paolo Favaro
    链接:https://arxiv.org/abs/1910.00287

    [跟踪相关]:

    【1】 Track to Reconstruct and Reconstruct to Track
    要重建的轨道和要重建的轨道
    作者: Jonathon Luiten, Bastian Leibe
    链接:https://arxiv.org/abs/1910.00130

    [迁移学习/domain/主动学习/自适应]:

    【1】 Harmonization of diffusion MRI datasets with adaptive dictionary learning
    扩散MRI数据集与自适应字典学习的协调
    作者: Samuel St-Jean, Alexander Leemans
    链接:https://arxiv.org/abs/1910.00272

    [裁剪/量化/加速相关]:

    【1】 Sub-Architecture Ensemble Pruning in Neural Architecture Search
    神经结构搜索中的子结构集成剪枝
    作者: Yijun Bian, Xia Hu
    链接:https://arxiv.org/abs/1910.00370

    [3D/3D重建等相关]:

    【1】 Towards Automatic Embryo Staging in 3D+T Microscopy Images using Convolutional Neural Networks and PointNets
    利用卷积神经网络和PointNets实现3D+T显微图像中的胚胎自动分期
    作者: Manuel Traub, Johannes Stegmaier
    链接:https://arxiv.org/abs/1910.00443

    【2】 DenseRaC: Joint 3D Pose and Shape Estimation by Dense Render-and-Compare
    DenseRaC:通过密集渲染和比较进行关节3D姿势和形状估计
    作者: Yuanlu Xu, Tony Tung
    备注:11 pages, 8 figures, International Conference on Computer Vision (ICCV) 2019, Oral Presentation
    链接:https://arxiv.org/abs/1910.00116

    [其他]:

    【1】 End-to-end learning of energy-based representations for irregularly-sampled signals and images
    用于不规则采样信号和图像的基于能量的表示的端到端学习
    作者: Ronan Fablet, François Rousseau
    链接:https://arxiv.org/abs/1910.00556

    【2】 A Three-dimensional Convolutional-Recurrent Network for Convective Storm Nowcasting
    用于对流风暴短时预报的三维对流-回归网络
    作者: Wei Zhang, Lei Han
    链接:https://arxiv.org/abs/1910.00527

    【3】 Graph convolutional networks for learning with few clean and many noisy labels
    用于学习的图卷积网络,具有很少的干净标签和许多噪声标签
    作者: Ahmet Iscen, Cordelia Schmid
    链接:https://arxiv.org/abs/1910.00324

    【4】 Underwhelming Generalization Improvements From Controlling Feature Attribution
    通过控制要素属性进行平淡无奇的概化改进
    作者: Joseph D. Viviano, Joseph Paul Cohen
    链接:https://arxiv.org/abs/1910.00199

    【5】 Custom Extended Sobel Filters
    自定义扩展Sobel过滤器
    作者: Victor Bogdan, Ciprian Orhei
    链接:https://arxiv.org/abs/1910.00138

    【6】 Hidden Trigger Backdoor Attacks
    隐藏触发后门攻击
    作者: Aniruddha Saha, Hamed Pirsiavash
    链接:https://arxiv.org/abs/1910.00033

    【7】 Map as The Hidden Sensor: Fast Odometry-Based Global Localization
    地图作为隐藏传感器:基于里程计的快速全球定位
    作者: Cheng Peng, David Weikersdorfer
    链接:https://arxiv.org/abs/1910.00572

    【8】 Leveraging Model Interpretability and Stability to increase Model Robustness
    利用模型的可解释性和稳定性提高模型的鲁棒性
    作者: Fei Wu, Alexandre Briot
    备注:Accepted at the 2019 ICCV workshop on Interpreting and Explaining Visual AI models; 8 pages
    链接:https://arxiv.org/abs/1910.00387

    【9】 Revisiting Fine-tuning for Few-shot Learning
    重新审视几次学习的微调
    作者: Akihiro Nakamura, Tatsuya Harada
    链接:https://arxiv.org/abs/1910.00216

    【10】 A Mobile Manipulation System for One-Shot Teaching of Complex Tasks in Homes
    一种用于家庭复杂任务一次性教学的移动操作系统
    作者: Max Bajracharya, Yutaka Takaoka
    备注:8 pages, 6 figures, submitted to IEEE Robotics and Automation Letters with option to present at 2020 Robotics International Conference on Robotics and Automation (ICRA)
    链接:https://arxiv.org/abs/1910.00127

    【11】 Risk-Aware Planning by Confidence Estimation using Deep Learning-Based Perception
    基于深度学习感知的基于置信度估计的风险感知规划
    作者: Maymoonah Toubeh, Pratap Tokekar
    链接:https://arxiv.org/abs/1910.00101

    【12】 Fitting IVIM with Variable Projection and Simplicial Optimization
    用变量投影和单纯形优化拟合IVIM
    作者: Shreyas Fadnavis, Eleftherios Garyfallidis
    链接:https://arxiv.org/abs/1910.00095

    【13】 Multi-Head Attention with Diversity for Learning Grounded Multilingual Multimodal Representations
    具有多样性的多头注意学习扎根多语言多模态表征
    作者: Po-Yao Huang, Alexander Hauptmann
    备注:Accepted at EMNLP 2019
    链接:https://arxiv.org/abs/1910.00058

    机器翻译,仅供参考

    相关文章

      网友评论

          本文标题:计算机视觉每日论文速递[10.02]

          本文链接:https://www.haomeiwen.com/subject/xvglpctx.html