# US 9,883,165 B2 · Method and system for reconstructing 3D trajectory in real time > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 9,883,165 B2 - Document ID: US9883165B2 - Status: Granted - Kind: B2 - Application number: 14/415,366 - Filed: 2013-07-10 - Published: 2015-07-23 - Granted: 2018-01-30 - Earliest claimed: 2012-07-16 - Assignees: Egidium Technologies - Applicants: Egidium Technologies - Inventors: Laurent Denizot; Fernand Alcatrao; Rodrigue Favrot; Cyril Sepieter - Web page: https://patentgrep.com/patent/US9883165B2 ## CPC - H04N 13/239: using two two-dimensional [2D] image sensors having a relative position equal to or related to the interocular distance (H04N13/243 takes precedence) - G06T 7/292: Multi-camera tracking - G06T 15/20: Perspective computation - G06T 19/00: Manipulating three-dimensional [3D] models or images for computer graphics - G06V 20/52: Surveillance or monitoring of activities, e.g. for recognising suspicious objects (recognising microscopic objects G06V20/69) - G08B 13/19608: {Tracking movement of a target, e.g. by detecting an object predefined as a target, using target direction and or velocity to predict its new position} - H04N 7/181: {for receiving images from a plurality of remote sources} - G06T 2207/30112: Baggage; Luggage; Suitcase - G06T 2207/30232: Surveillance - H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING - G08B: SIGNALLING SYSTEMS, e.g. PERSONAL CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS ## Abstract A method is for reconstructing a trajectory at a site by a processing server connected to cameras and having access to recordings from these cameras. The server implements a 3D visualization application based on virtual and augmented reality based on a static model of a 3D representation of the site, and a dynamic 3D representation database. The 3D visualization application is configured to determine, in the 3D visualization at a moment T0, the location of an object considered to have been left unattended by an owner, in response to a first instruction from the user, determine at least one camera which is best-positioned for displaying the object, access a recording from the camera to display the object at a moment corresponding to when the object was left unattended by an owner and activate a method for tracking the owner to determine the owner's current location. ## Claims ### Claim 1 A method for reconstructing a trajectory at a site by means of a processing server connected to a set of cameras and having access to the video recordings from these cameras, the method comprising: implementing a 3D visualization application by the processing server, wherein the 3D visualization application is based on virtual reality and augmented reality based on: a static model of a 3D representation of the site; and a dynamic 3D representation database; wherein the 3D visualization application is configured to: determine, in the 3D visualization at a moment T0, the location of an object considered to have been left unattended at the site by an owner, in response to a first instruction from the user; select at least one camera which is best-positioned for displaying said object, wherein the selection of the best-positioned camera comprises: determining three-dimensional coordinates of a point of impact of the owner with the ground based on a click made on a 2D monitor displaying a recording currently being viewed; calculating the three-dimensional coordinates of the point of impact of the owner with the ground in the 3D visualization based on a bijective function between the 3D visualization and the 2D monitor, and determining the best-positioned camera based on the calculated three-dimensional coordinates, access a video recording from the determined best-positioned camera so as to display the recording at a moment t prior to T0, t is determined in response to a second instruction corresponding to the moment when the object was left unattended by an owner; and activate a method for tracking the owner via the recordings from the different cameras from the moment t to the present moment so as to determine the current location of said owner. ### Claim 2 (depends on 1) The method according to claim 1, wherein the best-positionedcamera is determined by calculating a d/f ratio of each camera, d being the distance between the camera in question and the three-dimensional coordinates of the point of impact of the owner with the ground, f being the focal length of the camera in question. ### Claim 3 (depends on 2) The method according to claim 2, wherein the focal length and the position of the cameras remain unchanged over time. ### Claim 4 (depends on 1) The method according to claim 1, wherein the focal length is identical for all the cameras. ### Claim 5 (depends on 2) The method according to claim 2, wherein the processing server also recovers data relating to the dynamics of the focal length and orientation of the cameras so as to calculate the d/f ratio. ### Claim 6 (depends on 1) The method according to claim 1, wherein the 3D visualization and the monitors are integrated on the same display screen. ### Claim 7 (depends on 6) The method according to claim 6, wherein the display screen is of the touch-screen type. ### Claim 8 (depends on 1) The method according to claim 1, wherein, during the tracking method, the recordings are read at a speed greater than the normal recording speed. ### Claim 9 (depends on 1) The method according to claim 1, wherein, once the current location of the owner is determined, the owner's coordinates are transmitted wirelessly to remote mobile terminals. ### Claim 10 (depends on 1) The method according to claim 1, wherein the first instruction originates from a cursor capable of being moved by the user along a time line displayed in the 3D visualization. ### Claim 11 A system for reconstructing a trajectory at a site by means of a processing server connected to: a set of cameras and having access to the video recordings from these cameras, a static model of a 3D representation of the site, and a dynamic 3D representation database, said processing server implements a 3D visualization application based on virtual reality and augmented reality so as to: determine, in the 3D visualization at a moment T0, the location of an object considered to have been left unattended at the site by an owner, in response to a first instruction from the user; determine at least one camera which is best-positioned for displaying said object, wherein the selection of the best-positioned camera comprises: determining three-dimensional coordinates of a point of impact of the owner with the ground based on a click made on a 2D monitor displaying a recording currently being viewed; calculating the three-dimensional coordinates of the point of impact of the owner with the ground in the 3D visualization based on a bijective function between the 3D visualization and the 2D monitor, and determining the best-positioned camera based on the calculated three-dimensional coordinates, access a video recording from this camera so as to display the recording at a moment t prior to T0, t is determined in response to a second instruction corresponding to the moment when the object was left unattended by an owner; and activate a method for tracking the owner via the recordings from the different cameras from the moment t to the present moment so as to determine the current location of said owner.