US6035067A — Apparatus for tracking objects in video sequences…

Claim 1

Independent Claim 151 words

A method of image processing for tracking a subject in a sequence of image frames using defined criteria for classifying image regions as potentially being part of the tracked subject, the method comprising the steps of: (a) obtaining, for a first image frame, a record of an image region previously designated as containing the tracked subject; (b) classifying, in graded form, image regions of a second image frame in accordance with the defined criteria to thereby generate classified image regions, (c) defining a spatially varying threshold level with reference to the previously designated image region; and (d) updating said designated image region associated with said second image frame with reference to both said classified image regions in said second image frame and the previously designated image region in said first image frame, excluding image regions whose classification grade is below the spatially varying threshold level from the updated designated image region.

Claim 2

Independent Claim 151 words

A method of image processing for tracking a subject in a sequence of image frames using defined criteria for classifying image regions as potentially being part of the tracked subject, the method comprising the steps of: (a) obtaining, for a first image frame, a record of an image region previously designated as containing the tracked subject; (b) classifying, in graded form, image regions of a second image frame in accordance with the defined criteria to thereby generate classified image regions, (c) defining a spatially varying threshold level with reference to the previously designated image region; and (d) updating said designated image region associated with said second image frame with reference to both said classified image regions in said second image frame and the previously designated image region in said first image frame, excluding image regions whose classification grade is below the spatially varying threshold level from the updated designated image region.

Claim 3

Dependent Claim 40 words

The method as recited in claim 1, wherein said image regions classified as potentially part of the tracked subject are included or not included in the updated designated image region, depending on their proximity to the previously designated image region.

Claim 4

Dependent Claim 32 words

The method as recited in claim 1, wherein the updated designated image region is defined as a convex hull with reference to all of said image regions which are to be included.

Claim 5

Dependent Claim 37 words

The method as recited in claim 1, wherein the step of updating said designated image region is automatically inhibited to prevent excessive growth of said designated image region, said excessive growth being growth beyond a predetermined factor.

Claim 6

Dependent Claim 35 words

The method as recited in claim 4, wherein growth in area of said designated image region is judged excessive or not by comparison with an area value averaged over a number of previous image frames.

Claim 7

Dependent Claim 37 words

The method as recited in claim 1, wherein the step of updating said designated image region is automatically inhibited to prevent excessive shrinkage of said designated image region, said excessive shrinkage being shrinkage below a predetermined factor.

Claim 8

Dependent Claim 35 words

The method as recited in claim 6, wherein shrinkage in area of said designated image region is judged excessive or not by comparison with an area value averaged over a number of previous image frames.

Claim 9

Dependent Claim 57 words

The method as recited in claim 1, wherein the step of updating said designated image region is automatically inhibited in the event that the updated designated image region would include an excessive proportion of image regions not classified as potentially part of the tracked subject, said excessive proportion being a change in proportion beyond a predetermined factor.

Claim 10

Dependent Claim 66 words

The method as recited in claim 1, wherein motion of the tracked subject or designated image region is detected over successive image frames, and wherein updating of said designated image region in the updating step is performed with reference to the classified image regions in the second image frame, the previously designated image region and a predicted degree of motion with respect to the tracked subject.

Claim 11

Dependent Claim 28 words

The method as recited in claim 1, wherein each new image frame is divided into blocks of pixels and the classifying step is performed on a block-by-block basis.

Claim 12

Dependent Claim 38 words

The method as recited in claim 10, wherein the pixels of the first and second image frames are classified individually and the classification for each block is obtained by combining the classifications of the pixels within the block.

Claim 13

Independent Claim 33 words

The method as recited in 11, wherein individual pixels are classified in binary form and the classification of each block is perturbed by counting pixels within each block meeting a pixel classification criteria.

Claim 14

Independent Claim 33 words

The method as recited in 11, wherein individual pixels are classified in binary form and the classification of each block is perturbed by counting pixels within each block meeting a pixel classification criteria.

Claim 15

Independent Claim 17 words

The method as recited in claims 10, 11, or 12, wherein blocks are classified in binary form.

Claim 16

Dependent Claim 23 words

The method as recited in claim 1, wherein the defined criteria permit classification of moving objects as potentially part of the tracked subject.

Claim 17

Dependent Claim 45 words

The method as recited in claim 1, wherein the defined criteria for each of said image regions includes whether a pixel value or values of said image region in the second image frame is different from the corresponding pixel value of said first image frame.

Claim 18

Dependent Claim 21 words

The method as recited in claim 15, wherein for the classifying step, pixel differences below a predetermined noise threshold are disregarded.

Claim 19

Independent Claim 32 words

The method as claimed in claims 15 or 16, wherein the classifying step includes a substep of discriminating between changes representing movement of the tracked subject and changes representing uncovering of background.

Claim 20

Dependent Claim 32 words

The method as recited in claim 17, wherein: two successive comparisons are performed over three successive image frames, and changes are disregarded for image regions which do not change in both comparisons.

Claim 21

Dependent Claim 42 words

The method as recited in claim 1, wherein the classifying criteria for each of said image regions includes whether a pixel value or values of said image region in said second image frame fall within a predetermined subset of possible pixel values.

Claim 22

Dependent Claim 36 words

The method as recited in claim 19, wherein said predetermined subset is determined automatically by probability analysis of pixel values present in a designated image region when compared with said respective image frame as a whole.

Claim 23

Independent Claim 29 words

The method as recited in claims 19 or 20, wherein said predetermined subset of possible pixel values is defined with reference to the chrominance characteristics of the tracked subject.

Claim 24

Dependent Claim 24 words

The method as recited in claim 1, wherein a default designated image region is defined for the start of the sequence of image frames.

Claim 25

Dependent Claim 35 words

The method as recited in claim 22, wherein updating of the default designated image region is inhibited until a significant area within said designated image region is classified as potentially part of the tracked subject.

Claim 26

Dependent Claim 24 words

The method as recited in claim 1, wherein an initial designated image region for the sequence of image frames is defined by manual input.

Claim 27

Dependent Claim 26 words

The method as recited in claim 24, wherein the criteria used in the classifying step are defined automatically with reference to the initial designated image region.

Claim 28

Dependent Claim 60 words

The method as recited in claim 1, wherein said steps (a) through (d) comprise a routine for encoding motion picture image frames using a limited data rate, wherein image quality is enhanced for an image region designated as including the tracked subject, and wherein said designated image region is updated for the tracked subject using the steps (a) through (c).

Claim 29

Dependent Claim 30 words

The method as recited in claim 26, wherein said limited data rate is allocated on a block basis, and wherein said designated image region is updated in corresponding block units.

Claim 30

Independent Claim 15 words

The method as recited in claims 26 or 27, wherein the encoding is H261 compatible.

Claim 31

Independent Claim 15 words

The method as recited in claims 26 or 27, wherein the encoding is MPEG compatible.

Claim 32

Dependent Claim 25 words

The method as recited in claim 1, wherein the updating step is performed subject to a time delay defined with respect to a frame period.

Claim 33

Dependent Claim 60 words

The method as recited in claim 1, wherein steps (a) through (d) comprise a routine for encoding motion picture image frames using a limited data rate, wherein image quality is enhanced for the image region designated as including the particular tracked subject, and wherein said designated image region is updated for the tracked subject using the steps (a) through (d).

Claim 34

Dependent Claim 30 words

The method as recited in claim 31, wherein said limited data rate is allocated on a block basis, and wherein said designated image region is updated in corresponding block units.

Claim 35

Independent Claim 15 words

The method as recited in claims 31 or 32, wherein the encoding is H261 compatible.

Claim 36

Independent Claim 15 words

The method as recited in claims 31 or 32, wherein the encoding is MPEG compatible.

Claim 37

Dependent Claim 25 words

A videophone apparatus including a camera and means for encoding image frames output by the camera by the method recited in claims 26 or 31.

Claim 38

Dependent Claim 15 words

A motion picture signal generated by the method as recited in claims 26 or 31.

Claim 39

Dependent Claim 17 words

A motion picture recording conveying images encoded by a method as recited in claims 26 or 31.

Claim 40

Dependent Claim 92 words

A data processing apparatus wherein elements of a stored regular two-dimensional array are designated active or inactive, the apparatus comprising: means for identifying a convex hull encompassing all active elements of the stored array, said identifying means including: (a) means for scanning rows of the array to record the most extreme active element(s) in each row in a linear list; and (b) means for processing the linear list of extreme active elements to identify the desired convex hull without considering further the other less-extreme active elements in each row of the array.

Claim 41

Dependent Claim 92 words

A data processing apparatus wherein elements of a stored regular two-dimensional array are designated active or inactive, the apparatus comprising: means for identifying a convex hull encompassing all active elements of the stored array, said identifying means including: (a) means for scanning rows of the array to record the most extreme active element(s) in each row in a linear list; and (b) means for processing the linear list of extreme active elements to identify the desired convex hull without considering further the other less-extreme active elements in each row of the array.

Claim 42

Independent Claim 182 words

A method of image processing for tracking a subject in a sequence of image frames using a defined criteria for classifying image regions in said image frames as potentially being part of the tracked subject, the method comprising the steps of: (a) for a first image frame, designating an image region containing the tracked subject as a designated image region, (b) classifying image regions within said designated image region using the defined criteria; (c) obtaining a first record of said designated image region; (d) for a subsequent second image frame, obtaining a second record of said second image frame corresponding to the previously designated image region; (e) classifying, in graded form, image regions of said second image frame in accordance with the defined criteria; (f) defining a spatially varying threshold with reference to the previously designated image region, and (g) updating said designated image region with reference to both the classified image regions in said second image frame and to the first record, excluding image regions whose classification grade is below the spatially varying threshold level from the updated designated image region.

Claim 43

Independent Claim 182 words

A method of image processing for tracking a subject in a sequence of image frames using a defined criteria for classifying image regions in said image frames as potentially being part of the tracked subject, the method comprising the steps of: (a) for a first image frame, designating an image region containing the tracked subject as a designated image region, (b) classifying image regions within said designated image region using the defined criteria; (c) obtaining a first record of said designated image region; (d) for a subsequent second image frame, obtaining a second record of said second image frame corresponding to the previously designated image region; (e) classifying, in graded form, image regions of said second image frame in accordance with the defined criteria; (f) defining a spatially varying threshold with reference to the previously designated image region, and (g) updating said designated image region with reference to both the classified image regions in said second image frame and to the first record, excluding image regions whose classification grade is below the spatially varying threshold level from the updated designated image region.

Claim 44

Dependent Claim 40 words

The method as recited in claim 39, wherein said image regions classified as potentially part of the tracked subject are included or not included in the updated designated image region, depending on their proximity to the previously designated image region.

Claim 45

Dependent Claim 32 words

The method as recited in claim 39, wherein the updated designated image region is defined as a convex hull with reference to all of said image regions which are to be included.

Claim 46

Dependent Claim 37 words

The method as recited in claim 39, wherein the step of updating said designated image region is automatically inhibited to prevent excessive growth of said designated image region, said excessive growth being growth beyond a predetermined factor.

Claim 47

Dependent Claim 37 words

The method as recited in claim 39, wherein the step of updating said designated image region is automatically inhibited to prevent excessive shrinkage of said designated image region, said excessive shrinkage being shrinkage below a predetermined factor.

Claim 48

Dependent Claim 57 words

The method as recited in claim 39, wherein the step of updating said designated image region is automatically inhibited in the event that the updated designated image region would include an excessive proportion of image regions not classified as potentially part of the tracked subject, said excessive proportion being a change in proportion beyond a predetermined factor.

Claim 49

Dependent Claim 66 words

The method as recited in claim 39, wherein motion of the tracked subject or designated image region is detected over successive image frames, and wherein updating of said designated image region in the updating step is performed with reference to the classified image regions in the second image frame, the previously designated image region and a predicted degree of motion with respect to the tracked subject.

Claim 50

Dependent Claim 28 words

The method as recited in claim 39, wherein each new image frame is divided into blocks of pixels and the classifying step is performed on a block-by-block basis.

Claim 51

Dependent Claim 23 words

The method as recited in claim 39, wherein the defined criteria permit classification of moving objects as potentially part of the tracked subject.

Claim 52

Dependent Claim 45 words

The method as recited in claim 39, wherein the defined criteria for each of said image regions includes whether a pixel value or values of said image region in the second image frame is different from the corresponding pixel value of said first image frame.

Claim 53

Dependent Claim 42 words

The method as recited in claim 39, wherein the classifying criteria for each of said image regions includes whether a pixel value or values of said image region in said second image frame fall within a predetermined subset of possible pixel values.

Claim 54

Dependent Claim 24 words

The method as recited in claim 39, wherein a default designated image region is defined for the start of the sequence of image frames.

Claim 55

Dependent Claim 24 words

The method as recited in claim 39, wherein an initial designated image region for the sequence of image frames is defined by manual input.

Claim 56

Dependent Claim 61 words

The method as recited in claim 39, wherein said steps (a) through (f) comprise a routine for encoding motion picture image frames using a limited data rate, wherein image quality is enhanced for an image region designated as including a particular tracked subject, and wherein said designated image region is updated for the tracked subject using the steps (a) through (f).

Annuity Stage	Window Opens	Payment Deadline	Surcharge Period Until	Estimated Fee	Status
3.5 Years (Year 4)	2003-03-07	2003-09-06	2004-03-07	$2000.00	Window Closed / Lapsed
7.5 Years (Year 8)	2007-03-07	2007-09-06	2008-03-07	$3760.00	Window Closed / Lapsed
11.5 Years (Year 12)	2011-03-07	2011-09-06	2012-03-07	$7700.00	Window Closed / Lapsed

Apparatus for tracking objects in video sequences and methods therefor

Abstract

Inventors

Bibliographic Details

Claims Analysis

References Cited (Backward Citations)

USPTO Annuity & Maintenance Windows

Assigned Categories & Tags

Assign Category Tag