-
1.发明授权Systems and methods for image object recognition based on location information and object categories 有权基于位置信息和对象类别的图像对象识别系统和方法公开(公告)号：US09495619B2
公开(公告)日：2016-11-15
申请号：US14586033
申请日：2014-12-30
Applicant: Facebook, Inc.
Inventor： Kevin Dechau Tang , Lubomir Bourdev , Balamanohar Paluri , Robert D. Fergus
CPC classification number: G06K9/6256 , G06K9/00476 , G06K9/00671 , G06K9/42 , G06K9/4642 , G06K9/52 , G06K9/6267 , G06K9/627 , G06K9/66 , G06K2009/4666 , G06N3/08 , G06N3/084
Abstract: Systems, methods, and non-transitory computer-readable media can identify a set of regions corresponding to a geographical area. A collection of training images can be acquired. Each training image in the collection can be associated with one or more respective recognized objects and with a respective region in the set of regions. Histogram metrics for a plurality of object categories within each region in the set of regions can be determined based at least in part on the collection of training images. A neural network can be developed based at least in part on the histogram metrics for the plurality of object categories within each region in the set of regions and on the collection of training images.
Abstract translation: 系统，方法和非暂时的计算机可读介质可以识别对应于地理区域的一组区域。 可以获取训练图像的集合。 集合中的每个训练图像可以与一个或多个相应的识别对象以及区域集合中的相应区域相关联。 可以至少部分地基于训练图像的收集来确定该组区域中的每个区域内的多个对象类别的直方图度量。 可以至少部分地基于区域集合中的每个区域内的多个对象类别的直方图度量和训练图像的集合来开发神经网络。
-
Patent Agency Ranking
2.发明申请公开(公告)号：US20170300784A1
公开(公告)日：2017-10-19
申请号：US15640353
申请日：2017-06-30
Applicant: Facebook, Inc.
Inventor： Kevin Dechau Tang , Lubomir Bourdev , Balamanohar Paluri , Robert D. Fergus
CPC classification number: G06K9/6256 , G06K9/00476 , G06K9/00671 , G06K9/42 , G06K9/4642 , G06K9/52 , G06K9/6267 , G06K9/627 , G06K9/66 , G06K2009/4666 , G06N3/08 , G06N3/084
Abstract: Systems, methods, and non-transitory computer-readable media can identify a set of regions corresponding to a geographical area. A collection of training images can be acquired. Each training image in the collection can be associated with one or more respective recognized objects and with a respective region in the set of regions. Histogram metrics for a plurality of object categories within each region in the set of regions can be determined based at least in part on the collection of training images. A neural network can be developed based at least in part on the histogram metrics for the plurality of object categories within each region in the set of regions and on the collection of training images.
-
3.发明申请公开(公告)号：US20160189010A1
公开(公告)日：2016-06-30
申请号：US14586033
申请日：2014-12-30
Applicant: Facebook, Inc.
Inventor： Kevin Dechau Tang , Lubomir Bourdev , Balamanohar Paluri , Robert D. Fergus
CPC classification number: G06K9/6256 , G06K9/00476 , G06K9/00671 , G06K9/42 , G06K9/4642 , G06K9/52 , G06K9/6267 , G06K9/627 , G06K9/66 , G06K2009/4666 , G06N3/08 , G06N3/084
Abstract: Systems, methods, and non-transitory computer-readable media can identify a set of regions corresponding to a geographical area. A collection of training images can be acquired. Each training image in the collection can be associated with one or more respective recognized objects and with a respective region in the set of regions. Histogram metrics for a plurality of object categories within each region in the set of regions can be determined based at least in part on the collection of training images. A neural network can be developed based at least in part on the histogram metrics for the plurality of object categories within each region in the set of regions and on the collection of training images.
-
4.发明授权公开(公告)号：US10572771B2
公开(公告)日：2020-02-25
申请号：US15640353
申请日：2017-06-30
Applicant: Facebook, Inc.
Inventor： Kevin Dechau Tang , Lubomir Bourdev , Balamanohar Paluri , Robert D. Fergus
Abstract: Systems, methods, and non-transitory computer-readable media can identify a set of regions corresponding to a geographical area. A collection of training images can be acquired. Each training image in the collection can be associated with one or more respective recognized objects and with a respective region in the set of regions. Histogram metrics for a plurality of object categories within each region in the set of regions can be determined based at least in part on the collection of training images. A neural network can be developed based at least in part on the histogram metrics for the plurality of object categories within each region in the set of regions and on the collection of training images.
-
5.发明申请SYSTEMS AND METHODS FOR IMAGE OBJECT RECOGNITION BASED ON LOCATION INFORMATION AND OBJECT CATEGORIES 有权基于位置信息和对象类别的图像对象识别的系统和方法公开(公告)号：US20160342865A1
公开(公告)日：2016-11-24
申请号：US15229008
申请日：2016-08-04
Applicant: Facebook, Inc.
Inventor： Kevin Dechau Tang , Lubomir Bourdev , Balamanohar Paluri , Robert D. Fergus
CPC classification number: G06K9/6256 , G06K9/00476 , G06K9/00671 , G06K9/42 , G06K9/4642 , G06K9/52 , G06K9/6267 , G06K9/627 , G06K9/66 , G06K2009/4666 , G06N3/08 , G06N3/084
Abstract: Systems, methods, and non-transitory computer-readable media can identify a set of regions corresponding to a geographical area. A collection of training images can be acquired. Each training image in the collection can be associated with one or more respective recognized objects and with a respective region in the set of regions. Histogram metrics for a plurality of object categories within each region in the set of regions can be determined based at least in part on the collection of training images. A neural network can be developed based at least in part on the histogram metrics for the plurality of object categories within each region in the set of regions and on the collection of training images.
Abstract translation: 系统，方法和非暂时的计算机可读介质可以识别对应于地理区域的一组区域。 可以获取训练图像的集合。 集合中的每个训练图像可以与一个或多个相应的识别对象以及区域集合中的相应区域相关联。 可以至少部分地基于训练图像的收集来确定该组区域中的每个区域内的多个对象类别的直方图度量。 可以至少部分地基于区域集合中的每个区域内的多个对象类别的直方图度量和训练图像的集合来开发神经网络。
-
6.发明授权公开(公告)号：US09727803B2
公开(公告)日：2017-08-08
申请号：US15229008
申请日：2016-08-04
Applicant: Facebook, Inc.
Inventor： Kevin Dechau Tang , Lubomir Bourdev , Balamanohar Paluri , Robert D. Fergus
CPC classification number: G06K9/6256 , G06K9/00476 , G06K9/00671 , G06K9/42 , G06K9/4642 , G06K9/52 , G06K9/6267 , G06K9/627 , G06K9/66 , G06K2009/4666 , G06N3/08 , G06N3/084
Abstract: Systems, methods, and non-transitory computer-readable media can identify a set of regions corresponding to a geographical area. A collection of training images can be acquired. Each training image in the collection can be associated with one or more respective recognized objects and with a respective region in the set of regions. Histogram metrics for a plurality of object categories within each region in the set of regions can be determined based at least in part on the collection of training images. A neural network can be developed based at least in part on the histogram metrics for the plurality of object categories within each region in the set of regions and on the collection of training images.
-
-
-
-
-
Search Results
6
records
(took 0.014 seconds)
