US2022156973A1PendingUtilityA1

Information processing apparatus, information processing method, and program

Assignee: SONY GROUP CORPPriority: May 9, 2017Filed: Jan 31, 2022Published: May 19, 2022
Est. expiryMay 9, 2037(~10.8 yrs left)· nominal 20-yr term from priority
Inventors:Shunichi Homma
G06T 7/74G06T 7/70G06T 2207/20081G06T 7/73G06T 2207/20084G06T 2207/30244G06V 10/44
67
PatentIndex Score
0
Cited by
0
References
0
Claims

Abstract

To estimate a location and an attitude of a target object in real space in a more preferred aspect. Provided is an information processing apparatus including: an estimating unit configured to estimate at least one of a location or an attitude of a predetermined chassis in real space on the basis of a first image captured by a first image capturing unit among a plurality of image capturing units held in the chassis; and a verifying unit configured to verify a likelihood of the estimation result on the basis of a second image captured by a second image capturing unit having an optical axis different from an optical axis of the first image capturing unit among the plurality of image capturing units.

Claims

exact text as granted — not AI-modified
1 . An information processing apparatus, comprising:
 an estimating unit configured to estimate at least one of a location or an attitude of a chassis in real space based on a first image captured by a first image capturing unit of a plurality of image capturing units, wherein the plurality of image capturing units are held in the chassis; and   a verifying unit configured to verify a result of the estimation based on a comparison of a first feature amount extracted from a second image and a second feature amount, wherein
 the second image is captured by a second image capturing unit of the plurality of image capturing units, 
 the second image capturing unit has an optical axis different from an optical axis of the first image capturing unit, and 
 the second feature amount is acquired from a database. 
   
     
     
         2 . The information processing apparatus according to  claim 1 , wherein the database stores the second feature amount in association with at least one of the location or the attitude of the chassis in the real space. 
     
     
         3 . The information processing apparatus according to  claim 2 , wherein the second feature amount is acquired from the database based on the second image. 
     
     
         4 . The information processing apparatus according to  claim 3 , wherein
 the estimation unit is further configured to extract, from the second image, a first plurality of feature amounts as the first feature amount,
 wherein each feature amount of the first plurality of feature amounts corresponds to a respective feature point of a first plurality of feature points; and 
   the verifying unit is further configured to verify the result of the estimation by comparison of the extracted first plurality of feature amounts with a second plurality of feature amounts, wherein
 each feature amount of the second plurality of feature amounts corresponds to a respective feature point of a second plurality of feature points, and 
 the second plurality of feature amounts is stored in the database as the second feature amount. 
   
     
     
         5 . The information processing apparatus according to  claim 4 , wherein
 the estimation unit is further configured to extract, from the second image, a third plurality of feature amounts of first partial areas as the first feature amount,
 wherein the first partial areas include the first plurality of feature points; and 
   the verification unit is further configured to verify the result of the estimation by comparison of the extracted third plurality of feature amounts of the first partial areas with a fourth plurality of feature amounts of second partial areas,   wherein
 the second partial areas include the second plurality of feature points, and 
 the fourth plurality of feature amounts is stored in the database as the second feature amount. 
   
     
     
         6 . The information processing apparatus according to  claim 5 , wherein
 the estimation unit is further configured to calculate a similarity between each feature amount of the third plurality of feature amounts of the first partial areas and a corresponding feature amount of the fourth plurality of feature amounts of the second partial areas; and   the verification unit is further configured to verify the result of the estimation based on a number of feature points for which a result of the calculation of the similarity is equal to or greater than a threshold.   
     
     
         7 . The information processing apparatus according to  claim 3 , wherein
 at least two image capturing units of the plurality of image capturing units are set as a plurality of candidates for the second image capturing unit,   the second image capturing unit captures the second image as an extraction source,   the second feature amount is associated with a parameter based on the capture of the second image as the extraction source,   the parameter associated with the second feature amount is based on an attitude of the second image capturing unit in the real space,   the estimating unit is further configured to acquire the parameter for each of the plurality of candidates based on the first image, and   the verifying unit is further configured to:
 select at least one candidate of the plurality of candidates based on the parameter acquired for each candidate of the plurality of candidates and the parameter associated with the second feature amount, wherein the selected at least one candidate captures the second image; and 
 verify the result of the estimation based on the second image captured by the selected at least one candidate. 
   
     
     
         8 . The information processing apparatus according to  claim 1 , wherein
 the verification unit is further configured to verify the result of the estimation based on a discriminator generated by a machine learning process,   the machine learning process is based on the second image captured by the second image capturing unit, and   the discriminator is associated with at least one of the location or the attitude of the chassis in the real space.   
     
     
         9 . The information processing apparatus according to  claim 1 , wherein the verification unit is further configured to:
 acquire a fourth image captured by the second image capturing unit, wherein the capture of the fourth image is prior to capture of the second image by the second image capturing unit; and   verify the result of the estimation based on a similarity between the second image and the acquired fourth image.   
     
     
         10 . The information processing apparatus according to  claim 1 , wherein
 at least two image capturing units of the plurality of image capturing units correspond to the second image capturing unit,   each image capturing unit of the at least two image capturing units captures the second image, and   the verification unit is further configured to verify the result of the estimation based on the second image captured by each image capturing unit of the at least two image capturing units.   
     
     
         11 . The information processing apparatus according to  claim 1 , wherein the estimation unit is further configured to output the result of the estimation based on a first result of the verification. 
     
     
         12 . The information processing apparatus according to  claim 11 , wherein
 the estimation unit is further configured to:
 select a third image capturing unit from the plurality of image capturing units based on a second result of the verification different from the first result of the verification, wherein the third image capturing unit captures a third image; and 
 re-estimate at least one of the location or the attitude of the chassis in the real space based on the third image captured by the selected third image capturing unit. 
   
     
     
         13 . The information processing apparatus according to  claim 12 , wherein
 at least two image capturing units of the plurality of image capturing units are set as a plurality of candidates for the first image capturing unit, and   the estimation unit is further configured to select, based on the second result of the verification, the third image capturing unit from the at least two image capturing units that are set as the plurality of candidates for the first image capturing unit.   
     
     
         14 . The information processing apparatus according to  claim 12 , wherein
 the estimation unit is further configured to:
 select the first image capturing unit as a new second image capturing unit based on the second result of the verification; and 
 select the second image capturing unit as a new first image capturing unit based on the second result of the verification,
 wherein the selection of the first image capturing unit is before the selection of the second image capturing unit; and 
 
   the verification unit is further configured to verify the result of the estimation based on a fourth image captured by the new second image capturing unit.   
     
     
         15 . The information processing apparatus according to  claim 1 , wherein the chassis is a chassis of a mobile object, the first image capturing unit captures the first image in a direction different from a traveling direction of the mobile object, and
 the estimating unit is further configured to estimate at least one of the location or the attitude of the chassis in the real space based on the first image captured by the first image capturing unit.   
     
     
         16 . The information processing apparatus according to  claim 1 , wherein
 the second image capturing unit captures the second image in a direction opposite to an image capturing direction of the first image capturing unit, and   the verification unit is further configured to verify the result of the estimation based on the second image captured by the second image capturing unit.   
     
     
         17 . An information processing method, comprising:
 estimating at least one of a location or an attitude of a chassis in real space based on a first image captured by a first image capturing unit of a plurality of image capturing units, wherein the plurality of image capturing units are held in the chassis; and   verifying a result of the estimation based on a comparison of a first feature amount extracted from a second image and a second feature amount, wherein
 the second image is captured by a second image capturing unit of the plurality of image capturing units, 
 the second image capturing unit has an optical axis different from an optical axis of the first image capturing unit, and 
 the second feature amount is acquired from a database. 
   
     
     
         18 . A non-transitory computer-readable medium having stored thereon computer-executable instructions which, when executed by a processor, cause the processor to execute operations, the operations comprising:
 estimating at least one of a location or an attitude of a chassis in real space based on a first image captured by a first image capturing unit of a plurality of image capturing units, wherein the plurality of image capturing units are held in the chassis; and   verifying a result of the estimation based on a comparison of a first feature amount extracted from a second image and a second feature amount, wherein
 the second image is captured by a second image capturing unit of the plurality of image capturing units, 
 the second image capturing unit has an optical axis different from an optical axis of the first image capturing unit, and 
 the second feature amount is acquired from a database.

Join the waitlist — get patent alerts

Track US2022156973A1 — get alerts on status changes and closely related new filings.

We store only your email — no account needed. See our privacy policy.