Secure and private tracking across multiple cameras
First Claim
Patent Images
1. A system, comprising:
- computer-readable storage media having stored thereon instructions executable to cause the system to perform operations;
one or more hardware processors coupled to the computer-readable storage media and configured to read instructions from the computer-readable storage media media to cause the system to perform operations comprising;
obtaining by a remote server a plurality of images from a first video camera within a first privacy boundary, the plurality of images being provided to the remote server through a first communications network, the first privacy boundary being maintained by a secure configuration of the remote server, wherein the first privacy boundary prevents a user associated with the first privacy boundary from obtaining a plurality of images captured by a second video camera within a second privacy boundary;
detecting, by the remote server, an indication of at least one object in the plurality of video frames from the first video camera;
generating a first feature vector for the at least one object in the plurality of video frames from the first video camera;
receiving a request for a match evaluation for a second object at the remote server from an issuing computing system; and
determining if the at least one object in the plurality of video frames from the first video camera matches the second object using a distance function based upon the first feature vector and a second feature vector for the second object.
2 Assignments
0 Petitions
Accused Products
Abstract
A technique for multi-camera object tracking is disclosed that preserves privacy of imagery from each camera or group of cameras. This technique uses secure multi-party computation to compute a distance metric across data from multiple cameras without revealing any information to operators of the cameras except whether or not an object was observed by both cameras. This is achieved by a distance metric learning technique that reduces the computing complexity of secure computation while maintaining object identification accuracy.
-
Citations
20 Claims
-
1. A system, comprising:
-
computer-readable storage media having stored thereon instructions executable to cause the system to perform operations; one or more hardware processors coupled to the computer-readable storage media and configured to read instructions from the computer-readable storage media media to cause the system to perform operations comprising; obtaining by a remote server a plurality of images from a first video camera within a first privacy boundary, the plurality of images being provided to the remote server through a first communications network, the first privacy boundary being maintained by a secure configuration of the remote server, wherein the first privacy boundary prevents a user associated with the first privacy boundary from obtaining a plurality of images captured by a second video camera within a second privacy boundary; detecting, by the remote server, an indication of at least one object in the plurality of video frames from the first video camera; generating a first feature vector for the at least one object in the plurality of video frames from the first video camera; receiving a request for a match evaluation for a second object at the remote server from an issuing computing system; and determining if the at least one object in the plurality of video frames from the first video camera matches the second object using a distance function based upon the first feature vector and a second feature vector for the second object. - View Dependent Claims (2, 3, 4, 5, 6, 7)
-
-
8. A computer-implemented method, comprising:
-
obtaining by a remote server a plurality of images from a first video camera within a first privacy boundary, the plurality of images being provided to the remote server through a first communications network, the first privacy boundary being maintained by a secure configuration of the remote server, wherein the first privacy boundary prevents a user associated with the first privacy boundary from obtaining a plurality of images captured by a second video camera within a second privacy boundary; detecting, by the remote server, an indication of at least one object in the plurality of video frames from the first video camera; generating a first feature vector for the at least one object in the plurality of video frames from the first video camera; receiving a request for a match evaluation for a second object at the remote server from an issuing computing system; and determining if the at least one object in the plurality of video frames from the first video camera matches the second object using a distance function based upon the first feature vector and a second feature vector for the second object. - View Dependent Claims (9, 10, 11, 12, 13, 14, 20)
-
-
15. A non-transitory machine-readable medium having stored thereon machine-readable instructions executable to cause machine to perform operations comprising:
-
obtaining by a remote server a plurality of images from a first video camera within a first privacy boundary, the plurality of images being provided to the remote server through a first communications network, the first privacy boundary being maintained by a secure configuration of the remote server, wherein the first privacy boundary prevents a user associated with the first privacy boundary from obtaining a plurality of images captured by a second video camera within a second privacy boundary; detecting, by the remote server, an indication of at least one object in the plurality of video frames from the first video camera; generating a first feature vector for the at least one object in the plurality of video frames from the first video camera; receiving a request for a match evaluation for a second object at the remote server from an issuing computing system; and determining if the at least one object in the plurality of video frames from the first video camera matches the second object using a distance function based upon the first feature vector and a second feature vector for the second object. - View Dependent Claims (16, 17, 18, 19)
-
Specification