Patent US7489803B2 successfully refreshed from Google Patents.
US7489803B2
Expired / Lapsed
US
⏱️ 0.0 years remaining
Object detection
Assignee: Sony United Kingdom Ltd
•
Filing: 2004-12-08
•
Granted: 2009-02-10
•
Expires: 2024-12-08
Abstract
Abstract Object detection apparatus for detecting objects in a test image comprises means for comparing blocks of a test window of the image with reference data indicative of the presence of an object to generate index figures indicative of a degree of similarity between a region and the reference data; means for storing probability data corresponding to possible values of the index figure and the block position; means operable, in respect of a current block, for accessing a probability value from the storing means in dependence on that block's position within the test window and the index figure generated in respect of that block; and means for combining the probability values corresponding to blocks in a test window to generate a result indicative of the probability of that test window containing an object; in which the probability values in the storing means are ordered by block position then by index figure.
Inventors
- 👤 Simon Dominic Haynes
- 👤 Jonathan Living
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-09 12:33
CPC Classifications
G; G06; G06V; G06V40/00; G06V40/10; G06V40/16; G06V40/161; G06V40/165; G06F; G06F18/00; G06F18/20; G06F18/28; G06V10/00; G06V10/70; G06V10/77; G06V10/772; G06V40/167
Claims Analysis
Total Claims: 23
Claim 1
Independent Claim
159 words
Object detection apparatus configured to detect objects in a test image, said apparatus comprising: a comparator to compare blocks of a test window of said image with reference data indicative of the presence of an object to generate index figures indicative of a degree of similarity between a region and said reference data; storage to store probability data corresponding to possible values of said index figure and said block position; logic operable, in respect of a current block, to access a probability value from said storage in dependence on a position of the current block within said test window and said index figure generated in respect of the current block; and a combiner to combine said probability values corresponding to blocks in a test window to generate a result indicative of the probability of the test window including an object, wherein said probability values in said storage are ordered in the storage by block position, then by index figure.
Claim 2
Dependent Claim
159 words
Object detection apparatus configured to detect objects in a test image, said apparatus comprising: a comparator to compare blocks of a test window of said image with reference data indicative of the presence of an object to generate index figures indicative of a degree of similarity between a region and said reference data; storage to store probability data corresponding to possible values of said index figure and said block position; logic operable, in respect of a current block, to access a probability value from said storage in dependence on a position of the current block within said test window and said index figure generated in respect of the current block; and a combiner to combine said probability values corresponding to blocks in a test window to generate a result indicative of the probability of the test window including an object, wherein said probability values in said storage are ordered in the storage by block position, then by index figure.
Claim 3
Dependent Claim
44 words
Apparatus according to claim 1 , in which said accessing logic is further operable to access two or more probability values in respect of a position of a current block and index figure, said two or more probability values relating to different object orientations.
Claim 4
Dependent Claim
31 words
Apparatus according to claim 2 , in which said combiner is operable to combine probability values relating to a individual object orientations to generate a respective result for each object orientation.
Claim 5
Dependent Claim
27 words
Apparatus according to claim 2 , in which said probability values in said storage are ordered by object orientation, then by block position, then by index figure.
Claim 6
Dependent Claim
24 words
Apparatus according to claim 2 , in which said object orientations comprise at least a frontal orientation and an object directed to one side.
Claim 7
Dependent Claim
30 words
Apparatus according to claim 5 , in which said object orientations comprise at least a frontal orientation, an object directed to one side and an object directed to another side.
Claim 8
Dependent Claim
32 words
Apparatus according to claim 5 , in which probability values relating to said frontal orientation are stored in said storage to a higher resolution than probability values relating to other object orientations.
Claim 9
Dependent Claim
32 words
Apparatus according to claim 7 , in which probability values relating to said frontal orientation are stored in said storage with twice the resolution of probability values relating to other object orientations.
Claim 10
Dependent Claim
22 words
Apparatus according to claim 1 , in which said comparator is operable to project said blocks onto one or more image eigenvectors.
Claim 11
Dependent Claim
38 words
Apparatus according to claim 1 , in which said accessing logic further comprising: a cache store for caching recently accessed probability values and probability values near, in the storing order, to the probability values that were recently accessed
Claim 12
Dependent Claim
12 words
Apparatus according to claim 1 , in which said objects are faces.
Claim 13
Dependent Claim
10 words
Video conferencing apparatus comprising apparatus according to claim 1 .
Claim 14
Dependent Claim
9 words
Surveillance apparatus comprising apparatus according to claim 1 .
Claim 15
Dependent Claim
10 words
A camera arrangement comprising apparatus according to claim 1 .
Claim 16
Independent Claim
143 words
A method of detecting objects in a test image, said method comprising the steps of: comparing blocks of a test window of said image with reference data indicative of the presence of an object to generate index figures indicative of a degree of similarity between a region and said reference data; storing probability data corresponding to possible values of said index figure and said block position; accessing a probability value from the stored probability data in dependence on a position of a current block within said test window and the index figure generated in respect of the current block; and combining said probability values corresponding to blocks in a test window to generate a result indicative of the probability of the test window eentainiiig-including an object; wherein said stored probability values are ordered in a storage by block position, then by index figure.
Claim 17
Independent Claim
143 words
A method of detecting objects in a test image, said method comprising the steps of: comparing blocks of a test window of said image with reference data indicative of the presence of an object to generate index figures indicative of a degree of similarity between a region and said reference data; storing probability data corresponding to possible values of said index figure and said block position; accessing a probability value from the stored probability data in dependence on a position of a current block within said test window and the index figure generated in respect of the current block; and combining said probability values corresponding to blocks in a test window to generate a result indicative of the probability of the test window eentainiiig-including an object; wherein said stored probability values are ordered in a storage by block position, then by index figure.
Claim 18
Dependent Claim
30 words
The method according to claim 15 , wherein the probability values that are stored by said storing are ordered by object orientation, then by block position, then by index figure.
Claim 19
Dependent Claim
32 words
The method according to claim 15 further comprising the step of: caching recently accessed probability values and probability values near, in the storing order, to the probability values that were recently accessed.
Claim 20
Dependent Claim
151 words
A computer-readable medium encoded with computer code instructions that are configured, when executed by a computer, to perform a method including the steps of: comparing blocks of a test window of said image with reference data indicative of the presence of an object to generate index figures indicative of a degree of similarity between a region and said reference data; storing probability data corresponding to possible values of said index figure and said block position; accessing a probability value from the stored probability data in dependence on a position of a current block within said test window and the index figure generated in respect of the current block; combining said probability values corresponding to blocks in a test window to generate a result indicative of the probability of the test window including an object, wherein said stored probability values are ordered in a storage by block position, then by index figure.
Claim 21
Dependent Claim
151 words
A computer-readable medium encoded with computer code instructions that are configured, when executed by a computer, to perform a method including the steps of: comparing blocks of a test window of said image with reference data indicative of the presence of an object to generate index figures indicative of a degree of similarity between a region and said reference data; storing probability data corresponding to possible values of said index figure and said block position; accessing a probability value from the stored probability data in dependence on a position of a current block within said test window and the index figure generated in respect of the current block; combining said probability values corresponding to blocks in a test window to generate a result indicative of the probability of the test window including an object, wherein said stored probability values are ordered in a storage by block position, then by index figure.
Claim 22
Dependent Claim
31 words
The computer-readable medium according to claim 18 , wherein the probability values that are stored by said storing are ordered by object orientation, then by block position, then by index figure.
Claim 23
Dependent Claim
36 words
The computer-readable medium according to claim 18 , the method further comprising the step of: caching recently accessed probability values and probability values near, in the storing order, to the probability values that were recently accessed.
References Cited (Backward Citations)
EP0682325A2
Cited
US6681032B2
Cited
US7130454B1
Cited
US6496184B1
Cited
US20030215134A1
Cited
US6940545B1
Cited
US6829384B2
Cited
US20030190076A1
Cited
EP1353516A1
Cited
US7194114B2
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 2012-02-10 | 2012-08-11 | 2013-02-10 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2016-02-10 | 2016-08-11 | 2017-02-10 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2020-02-10 | 2020-08-11 | 2021-02-10 | $7700.00 | Window Closed / Lapsed |