专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明申请

US20170243084A1 DSP-SIFT: DOMAIN-SIZE POOLING FOR IMAGE DESCRIPTORS FOR IMAGE MATCHING AND OTHER APPLICATIONS 审中-公开
公开(公告)号：US20170243084A1
公开(公告)日：2017-08-24
申请号：US15345373
申请日：2016-11-07
申请人： THE REGENTS OF THE UNIVERSITY OF CALIFORNIA
发明人： Stefano Soatto , Jingming Dong
IPC分类号： G06K9/62 , G06K9/46 , G06K9/52
CPC分类号： G06K9/6267 , G06K9/4642 , G06K9/4671 , G06K9/52
摘要： A variation of scale-invariant feature transform (SIFT) based on pooling gradient orientations across different domain sizes, in addition to spatial locations. The resulting descriptor is called DSP-SIFT, and it outperforms other methods in wide-baseline matching benchmarks, including those based on convolutional neural networks, despite having the same dimension of SIFT and requiring no training. Problems of local representation of imaging data are also addressed as computation of minimal sufficient statistics that are invariant to nuisance variability induced by viewpoint and illumination. A sampling-based and a point-estimate based approximation of such representations are described.

2. 发明申请

US20190236399A1 VISUAL-INERTIAL SENSOR FUSION FOR NAVIGATION, LOCALIZATION, MAPPING, AND 3D RECONSTRUCTION 审中-公开
公开(公告)号：US20190236399A1
公开(公告)日：2019-08-01
申请号：US16059491
申请日：2018-08-09
申请人： THE REGENTS OF THE UNIVERSITY OF CALIFORNIA
发明人： Stefano Soatto , Konstantine Tsotsos
IPC分类号： G06K9/52 , G06T7/277 , G06T7/579 , G01S5/16 , G06K9/00 , G01C21/16
CPC分类号： G06K9/52 , G01C21/165 , G01S5/16 , G06K9/00664 , G06T7/277 , G06T7/579
摘要： A new method for improving the robustness of visual-inertial integration systems (VINS) based on derivation of optimal discriminants for outlier rejection, and the consequent approximations, that are both conceptually and empirically superior to other outlier detection schemes used in this context. It should be appreciated that VINS is central to a number of application areas including augmented reality (AR), virtual reality (VR), robotics, autonomous vehicles, autonomous flying robots, and so forth and their related hardware including mobile phones, such as for use in indoor localization (in GPS-denied areas), and the like.

3. 发明授权

US09418317B2 End-to end visual recognition system and methods 有权
标题翻译：端到端视觉识别系统和方法
公开(公告)号：US09418317B2
公开(公告)日：2016-08-16
申请号：US14245159
申请日：2014-04-04
申请人： THE REGENTS OF THE UNIVERSITY OF CALIFORNIA
发明人： Stefano Soatto , Taehee Lee
IPC分类号： G06K9/62 , G06K9/00 , G06K9/46 , G06T7/20
CPC分类号： G06K9/626 , G06K9/00671 , G06K9/00744 , G06K9/4671 , G06K9/6255 , G06T7/207 , G06T7/246
摘要： We describe an end-to-end visual recognition system, where “end-to-end” refers to the ability of the system of performing all aspects of the system, from the construction of “maps” of scenes, or “models” of objects from training data, to the determination of the class, identity, location and other inferred parameters from test data. Our visual recognition system is capable of operating on a mobile hand-held device, such as a mobile phone, tablet or other portable device equipped with sensing and computing power. Our system employs a video based feature descriptor, and we characterize its invariance and discriminative properties. Feature selection and tracking are performed in real-time, and used to train a template-based classifier during a capture phase prompted by the user. During normal operation, the system scores objects in the field of view based on their ranking.
摘要翻译：我们描述一种端到端的视觉识别系统，其中“端对端”是指系统执行系统的所有方面的能力，从构建场景的“地图”或“场景”的“模型” 从训练数据的对象，到测试数据的类，身份，位置等推测参数的确定。我们的视觉识别系统能够在配备有感测和计算能力的移动手持设备（例如移动电话，平板电脑或其他便携式设备）上操作。我们的系统采用基于视频的特征描述符，我们描述其不变性和辨别性质。功能选择和跟踪是实时执行的，用于在用户提示的捕获阶段训练基于模板的分类器。在正常操作期间，系统根据其排名对视场中的对象进行分数。

4. 发明申请

US20160140729A1 VISUAL-INERTIAL SENSOR FUSION FOR NAVIGATION, LOCALIZATION, MAPPING, AND 3D RECONSTRUCTION 审中-公开
标题翻译：用于导航，本地化，映射和3D重建的视觉传感器融合
公开(公告)号：US20160140729A1
公开(公告)日：2016-05-19
申请号：US14932899
申请日：2015-11-04
申请人： THE REGENTS OF THE UNIVERSITY OF CALIFORNIA
发明人： Stefano Soatto , Konstantine Tsotsos
IPC分类号： G06T7/20 , G06K9/52 , G01P15/18 , G06T7/00
CPC分类号： G06K9/52 , G01C21/165 , G01S5/16 , G06K9/00664 , G06T7/277 , G06T7/579
摘要： A new method for improving the robustness of visual-inertial integration systems (VINS) based on derivation of optimal discriminants for outlier rejection, and the consequent approximations, that are both conceptually and empirically superior to other outlier detection schemes used in this context. It should be appreciated that VINS is central to a number of application areas including augmented reality (AR), virtual reality (VR), robotics, autonomous vehicles, autonomous flying robots, and so forth and their related hardware including mobile phones, such as for use in indoor localization (in GPS-denied areas), and the like.
摘要翻译：一种基于派生离群反射抑制的最佳辨识方法，提高视觉 - 惯性积分系统（VINS）的鲁棒性的新方法，以及随后的近似，在概念上和经验上均优于本文中使用的其他异常值检测方案。应该理解，VINS是许多应用领域的核心，包括增强现实（AR），虚拟现实（VR），机器人，自主车辆，自主飞行机器人等及其相关硬件，包括移动电话，例如用于室内定位（在GPS拒绝区域）等。

5. 发明申请

US20140301635A1 END-TO-END VISUAL RECOGNITION SYSTEM AND METHODS 有权
标题翻译：端到端视觉识别系统和方法
公开(公告)号：US20140301635A1
公开(公告)日：2014-10-09
申请号：US14245159
申请日：2014-04-04
申请人： THE REGENTS OF THE UNIVERSITY OF CALIFORNIA
发明人： Stefano Soatto , Taehee Lee
IPC分类号： G06K9/62
CPC分类号： G06K9/626 , G06K9/00671 , G06K9/00744 , G06K9/4671 , G06K9/6255 , G06T7/207 , G06T7/246
摘要： We describe an end-to-end visual recognition system, where “end-to-end” refers to the ability of the system of performing all aspects of the system, from the construction of “maps” of scenes, or “models” of objects from training data, to the determination of the class, identity, location and other inferred parameters from test data. Our visual recognition system is capable of operating on a mobile hand-held device, such as a mobile phone, tablet or other portable device equipped with sensing and computing power. Our system employs a video based feature descriptor, and we characterize its invariance and discriminative properties. Feature selection and tracking are performed in real-time, and used to train a template-based classifier during a capture phase prompted by the user. During normal operation, the system scores objects in the field of view based on their ranking.
摘要翻译：我们描述一种端到端的视觉识别系统，其中“端对端”是指系统执行系统的所有方面的能力，从构建场景的“地图”或“场景”的“模型” 从训练数据的对象，到测试数据的类，身份，位置等推测参数的确定。我们的视觉识别系统能够在配备有感测和计算能力的移动手持设备（例如移动电话，平板电脑或其他便携式设备）上操作。我们的系统采用基于视频的特征描述符，我们描述其不变性和辨别性质。功能选择和跟踪是实时执行的，用于在用户提示的捕获阶段训练基于模板的分类器。在正常操作期间，系统根据其排名对视场中的对象进行分数。

6. 发明授权

US08717437B2 End-to end visual recognition system and methods 有权
标题翻译：端到端视觉识别系统和方法
公开(公告)号：US08717437B2
公开(公告)日：2014-05-06
申请号：US13735703
申请日：2013-01-07
申请人： The Regents of the University of California
发明人： Stefano Soatto , Taehee Lee
IPC分类号： H04N9/47
CPC分类号： G06K9/626 , G06K9/00671 , G06K9/00744 , G06K9/4671 , G06K9/6255 , G06T7/207 , G06T7/246
摘要： We describe an end-to-end visual recognition system, where “end-to-end” refers to the ability of the system of performing all aspects of the system, from the construction of “maps” of scenes, or “models” of objects from training data, to the determination of the class, identity, location and other inferred parameters from test data. Our visual recognition system is capable of operating on a mobile hand-held device, such as a mobile phone, tablet or other portable device equipped with sensing and computing power. Our system employs a video based feature descriptor, and we characterize its invariance and discriminative properties. Feature selection and tracking are performed in real-time, and used to train a template-based classifier during a capture phase prompted by the user. During normal operation, the system scores objects in the field of view based on their ranking.
摘要翻译：我们描述一种端到端的视觉识别系统，其中“端对端”是指系统执行系统的所有方面的能力，从构建场景的“地图”或“场景”的“模型” 从训练数据的对象，到测试数据的类，身份，位置等推测参数的确定。我们的视觉识别系统能够在配备有感测和计算能力的移动手持设备（例如移动电话，平板电脑或其他便携式设备）上操作。我们的系统采用基于视频的特征描述符，我们描述其不变性和辨别性质。功能选择和跟踪是实时执行的，用于在用户提示的捕获阶段训练基于模板的分类器。在正常操作期间，系统根据其排名对视场中的对象进行分数。

7. 发明申请

US20130215264A1 END-TO-END VISUAL RECOGNITION SYSTEM AND METHODS 有权
标题翻译：端到端视觉识别系统和方法
公开(公告)号：US20130215264A1
公开(公告)日：2013-08-22
申请号：US13735703
申请日：2013-01-07
申请人： THE REGENTS OF THE UNIVERSITY OF CALIFORNIA
发明人： Stefano Soatto , Taehee Lee
IPC分类号： G06K9/00
CPC分类号： G06K9/626 , G06K9/00671 , G06K9/00744 , G06K9/4671 , G06K9/6255 , G06T7/207 , G06T7/246
摘要： We describe an end-to-end visual recognition system, where “end-to-end” refers to the ability of the system of performing all aspects of the system, from the construction of “maps” of scenes, or “models” of objects from training data, to the determination of the class, identity, location and other inferred parameters from test data. Our visual recognition system is capable of operating on a mobile hand-held device, such as a mobile phone, tablet or other portable device equipped with sensing and computing power. Our system employs a video based feature descriptor, and we characterize its invariance and discriminative properties. Feature selection and tracking are performed in real-time, and used to train a template-based classifier during a capture phase prompted by the user. During normal operation, the system scores objects in the field of view based on their ranking.
摘要翻译：我们描述一种端到端的视觉识别系统，其中“端对端”是指系统执行系统的所有方面的能力，从构建场景的“地图”或“场景”的“模型” 从训练数据的对象，到测试数据的类，身份，位置等推测参数的确定。我们的视觉识别系统能够在配备有感测和计算能力的移动手持设备（例如移动电话，平板电脑或其他便携式设备）上操作。我们的系统采用基于视频的特征描述符，我们描述其不变性和辨别性质。功能选择和跟踪是实时执行的，用于在用户提示的捕获阶段训练基于模板的分类器。在正常操作期间，系统根据其排名对视场中的对象进行分数。

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式