US5471252A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Method and apparatus for estimating motion vector fields by rejecting local outliers
Assignee: Panasonic Corp of North America
•
Filing: 1993-11-03
•
Granted: 1995-11-28
•
Expires: 2013-11-03
Abstract
Abstract An image analysis system generates a motion vector field from first and second images in a sequence of steps. First, random, initial motion vector values are assigned to each pixel in the image frame. Next, an objective function is defined for each motion vector and its corresponding motion vector in the other frame. The differences between the motion vectors of the target pixel value in the current frame and the other frame are minimized by minimizing the objective function values, first based on all neighboring pixel values. Next, the motion vector differences are further minimized by reducing the objective function values by rejecting motion vectors which correspond to the neighboring pixel values for which the value of the difference between the respective objective functions is outside of a threshold range. In one embodiment, pixel values in the first image frame are compared to neighboring pixel values in a second image frame which occurs after the first image and to other neighboring pixel values in a third image frame which occurs before the first image frame. This embodiment substantially eliminates motion vector errors due to both motion discontinuities and occlusion of pixels in the image.
Inventors
- 👤 Siu-Leong Iu
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-08 14:03
CPC Classifications
G; G06; G06T; G06T7/00; G06T7/20; G06T7/269
Claims Analysis
Total Claims: 20
Claim 1
Independent Claim
292 words
Apparatus for generating sample values representing a motion vector field which describes motion of individual image components of a current image frame and corresponding image components of at least one other image frame in a sequence of image frames, the apparatus comprising: memory means for holding data values representing respective image components of the current image frame and the one other image frame; initialization means, coupled to the memory means, for generating a plurality of motion vectors, each of said plurality of motion vectors corresponding to an estimate of motion between one of the image components in the current image frame and a corresponding image component in the one other image frame; means for generating an energy function value for a target motion vector of the plurality of motion vectors which said target motion vector corresponds to a current image component in the current image frame, said energy function value being a function of the target motion vector and of neighboring motion vectors which said neighboring motion vectors are ones of the plurality of motion vectors which correspond to ones of the image components that surround the current image component in the current image frame; means for selecting ones of said neighboring motion vectors of the target motion vector for further processing, each one of the neighboring motion vectors is selected if the one neighboring motion vector differs in value from the target motion vector by less than a predetermined threshold; and means for minimizing the energy function value of each motion vector of the plurality of motion vectors as a function of the selected motion vectors to produce a corresponding optimized motion vector value, wherein the optimized motion vectors corresponding to the plurality of motion vectors define the motion vector field.
Claim 2
Dependent Claim
292 words
Apparatus for generating sample values representing a motion vector field which describes motion of individual image components of a current image frame and corresponding image components of at least one other image frame in a sequence of image frames, the apparatus comprising: memory means for holding data values representing respective image components of the current image frame and the one other image frame; initialization means, coupled to the memory means, for generating a plurality of motion vectors, each of said plurality of motion vectors corresponding to an estimate of motion between one of the image components in the current image frame and a corresponding image component in the one other image frame; means for generating an energy function value for a target motion vector of the plurality of motion vectors which said target motion vector corresponds to a current image component in the current image frame, said energy function value being a function of the target motion vector and of neighboring motion vectors which said neighboring motion vectors are ones of the plurality of motion vectors which correspond to ones of the image components that surround the current image component in the current image frame; means for selecting ones of said neighboring motion vectors of the target motion vector for further processing, each one of the neighboring motion vectors is selected if the one neighboring motion vector differs in value from the target motion vector by less than a predetermined threshold; and means for minimizing the energy function value of each motion vector of the plurality of motion vectors as a function of the selected motion vectors to produce a corresponding optimized motion vector value, wherein the optimized motion vectors corresponding to the plurality of motion vectors define the motion vector field.
Claim 3
Dependent Claim
18 words
Apparatus according to claim 1, wherein each image component is an individual picture element (pixel) of the image.
Claim 4
Dependent Claim
17 words
Apparatus according to claim 1, wherein each image component includes a plurality of individual picture elements (pixels).
Claim 5
Dependent Claim
49 words
Apparatus according to claim 1, wherein the means for generating an energy function value for the target motion vector includes means for rejecting an outlier motion vector as a function of only the neighboring motion vectors which are adjacent to the target motion vector in the motion vector field.
Claim 6
Dependent Claim
78 words
Apparatus according to claim 4, wherein the means for generating the energy function value includes: means for comparing each of N c neighboring motion vectors which surround the target motion vector, where N c is a positive integer, to identify a predetermined number of the N c neighboring motion vectors which have larger magnitudes than any other ones of the N c neighboring motion vectors; and means for rejecting the identified neighboring motion vectors as outlier motion vectors.
Claim 7
Dependent Claim
98 words
Apparatus according to claim 4, wherein the means for generating the energy function value includes: means for evaluating each of N c neighboring motion vectors which surround the target motion vector, where N c is a positive integer, to assign a magnitude value to each of the N c neighboring motion vectors; means for identifying one of the assigned magnitude values as being a median magnitude value for the N c neighboring motion vectors; and means for rejecting the identified neighboring motion vectors which have magnitude values greater than the median magnitude value as being outlier motion vectors.
Claim 8
Dependent Claim
82 words
Apparatus according to claim 4, wherein the means for minimizing the energy function value is an iterative process which includes means for switching the means for generating the energy function value, at a predetermined instant in the iterative process, from using a first energy function to generate the energy function value, which said first energy function uses all neighboring motion vectors, to use a second energy function to generate the energy function value, which said second energy function rejects outlier motion vectors.
Claim 9
Dependent Claim
253 words
Apparatus according to claim 1, wherein the motion vector field describes the motion of individual image components of the current image frame and corresponding image components in first and second image frame which differ from the current image frame, wherein: the initialization means includes means for generating a first plurality of motion vectors, each of said first plurality of motion vectors corresponding to an estimate of motion between one of the image components in the current image frame and a corresponding image component in the first image frame, and for generating a second plurality of motion vectors, each of said second plurality of motion vectors corresponding to an estimate of motion between one of the image components in the current image frame and a corresponding image component in the second image frame; and the means for generating an energy function value for the target motion vector includes: means for generating a first energy function value for the target motion vector, the first energy function value being a function of the target motion vector and respective neighboring ones of the first plurality of motion vectors; means for generating a second energy function value for the target motion vector, the second energy function value being a function of the target motion vector and respective neighboring ones of the second plurality of motion vectors; and means for choosing, as the generated energy function value, one of the first and second energy function values based on differences in magnitude between the first and second energy function values.
Claim 10
Dependent Claim
101 words
Apparatus according to claim 8, wherein the means for generating an energy function value for the target motion vector includes: means for generating a third energy function value for the target motion vector, the third energy function value being a function of the target motion vector, the neighboring ones of the first plurality of motion vectors and the neighboring ones of the second plurality of motion vectors; and means for choosing, as the generated energy function value, one of the first, second and third energy function values based on differences in magnitude among the first, second and third energy function values.
Claim 11
Dependent Claim
35 words
Apparatus according to claim 1, wherein the means for minimizing the energy function value in the current image frame includes means for processing a plurality of the energy function values using a simulated annealing algorithm.
Claim 12
Independent Claim
260 words
A method of generating sample values representing a motion vector field which describes motion of individual image components of a current image frame and corresponding image components of at least one other image frame in a sequence of image frames, the method comprising the steps of: a) holding data values representing respective image components of the current image frame in a memory; b) generating a plurality of motion vectors, each of the plurality of motion vectors corresponding to a respectively different one of the image components in the current image frame; c) generating an energy function value for a target motion vector of the plurality of motion vectors which said target motion vector corresponds to a current image component in the current image frame, said energy function value being a function of the target motion vector and of neighboring motion vectors which said neighboring motion vectors are ones of the plurality of motion vectors which correspond to ones of the image components that surround the current image component in the current image frame; d) selecting ones of said neighboring motion vectors of the target motion vector for further processing, each one of the neighboring motion vectors is selected if the one neighboring motion vector differs in value from the target motion vector by less than a predetermined threshold; and e) minimizing the energy function value of vectors as a function of the selected motion vectors to produce a corresponding optimized motion vector value, wherein the optimized motion vectors corresponding to the plurality of motion vectors define the motion vector field.
Claim 13
Independent Claim
260 words
A method of generating sample values representing a motion vector field which describes motion of individual image components of a current image frame and corresponding image components of at least one other image frame in a sequence of image frames, the method comprising the steps of: a) holding data values representing respective image components of the current image frame in a memory; b) generating a plurality of motion vectors, each of the plurality of motion vectors corresponding to a respectively different one of the image components in the current image frame; c) generating an energy function value for a target motion vector of the plurality of motion vectors which said target motion vector corresponds to a current image component in the current image frame, said energy function value being a function of the target motion vector and of neighboring motion vectors which said neighboring motion vectors are ones of the plurality of motion vectors which correspond to ones of the image components that surround the current image component in the current image frame; d) selecting ones of said neighboring motion vectors of the target motion vector for further processing, each one of the neighboring motion vectors is selected if the one neighboring motion vector differs in value from the target motion vector by less than a predetermined threshold; and e) minimizing the energy function value of vectors as a function of the selected motion vectors to produce a corresponding optimized motion vector value, wherein the optimized motion vectors corresponding to the plurality of motion vectors define the motion vector field.
Claim 14
Dependent Claim
19 words
A method according to claim 11, wherein each image component is an individual picture element (pixel) of the image.
Claim 15
Dependent Claim
18 words
A method according to claim 11, wherein each image component includes a plurality of individual picture elements (pixels).
Claim 16
Dependent Claim
38 words
A method according to claim 11, wherein step c) includes the step of rejecting an outlier motion vector based only on the neighboring motion vectors which are adjacent to the target motion vector in the motion vector field.
Claim 17
Dependent Claim
62 words
A method according to claim 14, wherein step f) is an iterative process which includes the step of switching, at a predetermined instant in the iterative process, from using a first energy function, which uses all neighboring motion vectors, to generate the energy function value to using a second energy function, which rejects outlier motion vectors, to generate the energy function value.
Claim 18
Dependent Claim
237 words
A method according to claim 11, wherein the motion vector field describes the motion of individual image components of the current image frame and corresponding image components in first and second image frames which differ from the current image frame, wherein: step b) includes the steps of generating a first plurality of motion vectors, each of the first plurality of motion vectors corresponding to an estimate of motion between one of the pixels in the current image frame and a corresponding pixel in the first image frame, and generating a second plurality of motion vectors, each of the second plurality of motion vectors corresponding to an estimate of motion between one of the image components in the current image frame and a corresponding image component in the second image frame; and step c) includes the steps of: generating a first energy function value for the target motion vector, the first energy function value being a function of the target motion vector and respective neighboring ones of the first plurality of motion vectors; generating a second energy function value for the target motion vector, the second energy function value being a function of the target motion vector and respective neighboring ones of the second plurality of motion vectors; and choosing, as the generated energy function value, one of the first and second energy function values based on differences in magnitude between the first and second energy function values.
Claim 19
Dependent Claim
101 words
A method according to claim 16, wherein the step of generating an energy function value for the target motion vector includes the steps of: generating a third energy function value for the target motion vector, the third energy function value being a function of the target motion vector, the neighboring ones of the first plurality of motion vectors and the neighboring ones of the second plurality of motion vectors; and choosing, as the generated energy function value, one of the first, second and third energy function values based on differences in magnitude among the first, second and third energy function values.
Claim 20
Dependent Claim
26 words
A method according to claim 11, wherein step e) includes the step of processing a plurality of the energy function values using a simulated annealing algorithm.
References Cited (Backward Citations)
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 1998-11-27 | 1999-05-29 | 1999-11-28 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2002-11-27 | 2003-05-29 | 2003-11-28 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2006-11-27 | 2007-05-29 | 2007-11-28 | $7700.00 | Window Closed / Lapsed |