EXTENDING DATA-DRIVEN DETECTION TO THE PREDICTION OF OBJECT PART LOCATIONS

• US 20150278632A1
• Filed: 03/28/2014
• Published: 10/01/2015
• Est. Priority Date: 03/28/2014
• Status: Active Grant

• Current Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Original Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Inventors: Rodriguez-Serrano, Jose-Antonio, Dai, Zhenwen
• Granted Patent: US 9,842,274 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/583 using metadata automatically derived from the content
• CPC Class Codes: G06F 16/5854 using shape and object relationship
• CPC Class Codes: G06F 18/214 Generating training patterns; Bootstrap methods, e.g. bagging or boosting
• CPC Class Codes: G06F 18/22 Matching criteria, e.g. proximity measures
• CPC Class Codes: G06V 10/25 Determination of region of interest [ROI] or a volume of interest [VOI]
• CPC Class Codes: G06V 20/54 of traffic, e.g. cars on the road, trains or boats

Abstract

Methods and systems for detecting an object part location based on an extended date-driven detection. A specific relevance value between configurations of parts with respect to a set of training images annotated with annotating objects can be defined. A similarity learning algorithm can be applied with respect to the parts to obtain a similarity function based on the similarity between the part configurations. The similarity learning algorithm receives a set of positive pair having similar part configuration and a negative pair having different configuration and returns the similarity function that tends to assign a high score to new positive pair and a low score to negative pairs. A similarity between a new image and the training images can be measured utilizing the learned similarity function to obtain a neighbor image and a visible and/or non-visible part location with respect to the image can be predicted based on the neighbor image.

Claims

1. A method for detecting an object part location, said method comprising:

defining a relevance value between a configuration of a plurality of parts and a set of training images annotated via an annotating object;

applying a similarity learning algorithm with respect to said plurality of parts to obtain a learned similarity function based on said relevance value; and

measuring a similarity between a new image and said set of training images utilizing said learned similarity function in order to obtain a neighbor image and predict a visible and/or a non-visible part location with respect to said new image based on said neighbor image.

2. The method of claim 1 further comprising:

receiving a set of positive pairs having similar part configurations and a negative pair having different part configurations by said similarity learning algorithm; and

returning said similarity function to assign a high score to said positive pair and a low score to said negative pair.

3. The method of claim 1 further comprising capturing said training images and said new image with respect to said object by an image capturing unit and processing said image utilizing an image processing unit.

4. The method of claim 1 wherein said annotating object with respect to said training images and said new image comprises a rectangular shape having at least four coordinates.

5. The method of claim 4 wherein said annotating object comprises an empty value with respect to said invisible object part location.

6. The method of claim 1 further comprising:

defining a set of indices corresponding to said visible part in said training images and thereafter compute said relevance value between said training images;

computing a part-by-part overlap score by measuring an overlap between said similar parts; and

assigning said relevance value to one if said part-by-part overlap score is greater than a threshold otherwise assign zero.

7. The method of claim 6 wherein said threshold depends on constraints with respect to a data driven detection application.

8. The method of claim 1 further comprising assigning said relevance value to zero if said different parts are visible in said training images.

9. The method of claim 1 further comprising:

outputting said object part location with respect to said new image for a single neighbor; and

combining said object part location in case of a plurality of neighbours.

10. The method of claim 1 further comprising:

computing a weighted average for said visible part with respect to said plurality of neighbors; and

returning an empty value with respect to said plurality of neighbor having said invisible part.

11. A system for detecting an object part location, said system comprising:

a processor; and

a computer-usable medium embodying computer program code, said computer-usable medium capable of communicating with the processor, said computer program code comprising instructions executable by said processor and configured for;

defining a relevance value between a configuration of a plurality of parts and a set of training images annotated via an annotating object;

applying a similarity learning algorithm with respect to said plurality of parts to obtain a learned similarity function based on said relevance value; and

measuring a similarity between a new image and said set of training images utilizing said learned similarity function in order to obtain a neighbour image and predict a visible and/or a non-visible part location with respect to said new image based on said neighbour image.

12. The system of claim 11 wherein said instructions are further configured for:

receiving a set of positive pairs having similar part configurations and a negative pair having different part configurations by said similarity learning algorithm; and

returning said similarity function to assign a high score to said positive pair and a low score to said negative pair.

13. The system of claim 11 wherein said instructions are further configured for capturing said training images and said new image with respect to said object by an image capturing unit and processing said image utilizing an image processing unit.

14. The system of claim 11 wherein said annotating the object with respect to said training images and said new image comprises a rectangular shape having at least four coordinates.

15. The system of claim 14 wherein said annotating the object comprises an empty value with respect to said invisible object part location.

16. A processor-readable medium storing code representing instructions to cause a process for detecting an object part location, said code comprising code to:

define a relevance value between a configuration of a plurality of parts and a set of training images annotated via an annotating object;

apply a similarity learning algorithm with respect to said plurality of parts to obtain a learned similarity function based on said relevance value; and

measure a similarity between a new image and said set of training images utilizing said learned similarity function in order to obtain a neighbour image and predict a visible and/or a non-visible part location with respect to said new image based on said neighbour image.

17. The processor-readable medium of claim 16 wherein said code further includes code to:

receive a set of positive pairs having similar part configurations and a negative pair having different part configurations by said similarity learning algorithm; and

return said similarity function to assign a high score to said positive pair and a low score to said negative pair.

18. The processor-readable medium of claim 16 wherein said code further includes code to capture said training images and said new image with respect to said object by an image capturing unit and processing said image utilizing an image processing unit.

19. The processor-readable medium of claim 16 wherein said annotating the object with respect to said training images and said new image comprises a rectangular shape having at least four coordinates.

20. The processor-readable medium of claim 19 wherein said annotating the object comprises an empty value with respect to said invisible object part location.