# US 8,542,268 B2 · Compact multi-aperture three-dimensional imaging system > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 8,542,268 B2 - Document ID: US8542268B2 - Status: Granted - Kind: B2 - Application number: 13/039,052 - Filed: 2011-03-02 - Published: 2012-09-06 - Granted: 2013-09-24 - Earliest claimed: 2011-03-02 - Assignees: Ricoh Company, Ltd. - Applicants: Sri Rama Prasanna Pavani; Jorge Moraleda; David G. Stork; Kathrin Berkner - Inventors: Sri Rama Prasanna Pavani; Jorge Moraleda; David G. Stork; Kathrin Berkner - Web page: https://patentgrep.com/patent/US8542268B2 ## CPC - H04N 13/218: using spatial multiplexing - H04N 13/271: wherein the generated image signals comprise depth maps or disparity maps - H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION ## Abstract A stereo-imaging system estimates depth based on two images; a multiview-imaging system estimates depth based on two or more images. A passive multiview-imaging system based on a sensor array is designed by considering different candidate partitions of the sensor array into imaging subsystems, especially three or more imaging subsystems. ## Claims ### Claim 1 A method for designing a passive multiview-imaging system based on a sensor array and a processor, the sensor array divided into partitions, each partition forming a sensor area for a corresponding imaging subsystem, each imaging subsystem capturing an image of an object, the processor estimating depth in the object based on the captured images, the design method comprising a computer system performing the steps of: automatically generating a plurality of different candidate partitions, each candidate partition defining a division of the sensor array into two or more sensor areas, wherein at least some of the candidate partitions divide the sensor array into three or more sensor areas; estimating a quality metric for each candidate partition, comprising: accessing an optical design for the corresponding imaging subsystems for the candidate partition; modeling the image capture by the imaging subsystems and the depth estimation by the processor, the modeling taking into account the input specifications; and estimating the quality metric based on the modeled depth estimation; and ordering the candidate partitions according to their estimated quality metrics. ### Claim 2 (depends on 1) The method of claim 1 wherein all of the automatically generated candidate partitions divide the sensor array into a regular array of sensor areas. ### Claim 3 (depends on 2) The method of claim 2 wherein all of the automatically generated candidate partitions divide the sensor array into a rectangular array of sensor areas. ### Claim 4 (depends on 1) The method of claim 1 wherein at least one of the automatically generated candidate partitions divides the sensor array in a manner that is not a regular array of sensor areas. ### Claim 5 (depends on 1) The method of claim 1 wherein the step of the computer system automatically generating a plurality of different candidate partitions generates only candidate partitions that are consistent with manufacturability constraints in the input specifications. ### Claim 6 (depends on 1) The method of claim 1 wherein a majority of the candidate partitions divide the sensor array into three or more sensor areas. ### Claim 7 (depends on 1) The method of claim 1 wherein not more than one of the candidate partitions divides the sensor array into two sensor areas. ### Claim 8 (depends on 1) The method of claim 1 wherein the quality metric comprises an error in a three-dimensional position estimation. ### Claim 9 (depends on 1) The method of claim 1 wherein the quality metric comprises an uncertainty in a three-dimensional position estimation. ### Claim 10 (depends on 1) The method of claim 1 wherein the step of estimating the quality metric comprises: using a closed form solution to estimate a three-dimensional position (X,Y,Z) of an object. ### Claim 11 (depends on 1) The method of claim 1 wherein the step of estimating the quality metric comprises: using a Monte Carlo approach to estimate uncertainty in a three-dimensional position (X,Y,Z) of an object. ### Claim 12 (depends on 1) The method of claim 1 wherein the step of the computer system estimating the quality metric comprises: estimating a disparity of the candidate partition; and estimating the quality metric based on the disparity. ### Claim 13 (depends on 1) The method of claim 1 wherein the optical designs for all of the imaging subsystems are the same. ### Claim 14 (depends on 1) The method of claim 1 wherein the optical designs for all of the imaging subsystems are variations of a common base optical design. ### Claim 15 (depends on 1) The method of claim 1 wherein, for at least two of the imaging subsystems, their corresponding optical designs have different spectral responses, different transfer functions and/or different polarization responses. ### Claim 16 (depends on 1) The method of claim 1 wherein the step of estimating a quality metric for each candidate partition further comprises: the computer system modifying the depth estimation by the processor to improve the quality metric. ### Claim 17 (depends on 1) The method of claim 1 wherein the step of ordering the candidate partitions comprises: the computer system identifying the candidate partition with the best quality metric. ### Claim 18 (depends on 1) The method of claim 1 wherein the step of ordering the candidate partitions comprises: the computer system rank ordering the candidate partitions by quality metric. ### Claim 19 (depends on 1) The method of claim 1 wherein the step of ordering the candidate partitions comprises: the computer system presenting the quality metric as a function of an aspect of the candidate partitions. ### Claim 20 A passive polyview-imaging system for estimating depth in an object, the polyview-imaging system comprising: a sensor array, the sensor array divided into at least two sensor areas; a plurality of imaging subsystems, each imaging subsystem comprising one of the sensor areas and corresponding optics, each imaging subsystem capturing an image of an object; and a processor coupled to the sensor array, for estimating depth in the object based on the captured images; wherein the passive polyview-imaging system is produced by a process comprising a computer system performing the steps of: automatically generating a plurality of different candidate partitions, each candidate partition defining a division of the sensor array into two or more sensor areas; estimating a quality metric for each candidate partition, comprising: accessing an optical design for the corresponding imaging subsystems for the candidate partition; modeling the image capture by the imaging subsystems and the depth estimation by the processor; and estimating the quality metric based on the modeled depth estimation; and ordering the candidate partitions according to their estimated quality metrics.