US5557684A
Expired / Lapsed
US
⏱️ 0.0 years remaining
System for encoding image data into multiple layers representing regions of coherent motion and associated motion parameters
Assignee: Massachusetts Institute of Technology
•
Filing: 1994-12-27
•
Granted: 1996-09-17
•
Expires: 2014-12-27
Abstract
Abstract A system stores images as a series of layers by determining (i) the boundaries of regions of coherent motion over the entire image, or frame, sequence; and (ii) associated motion parameters, or coefficients of motion equations, that describe the transformations of the regions from frame to frame. The system first estimates motion locally, by determining the movements within small neighborhoods of pixels from one image frame i to the next image frame i+1, to develop an optical flow, or dense motion, model of the image. Next, the system estimates the motion using affine or other low order, smooth transformations within a set of regions which the system has previously identified as having coherent motion, i.e., identified by analyzing the motions in the frames i-1 and i. It groups, or clusters, similar motion models and iteratively produces an updated set of models for the image. The system then uses the local motion estimates to associate individual pixels in the image with the motion model that most closely resembles the pixel's movement, to update the regions of coherent motion. Using these updated regions, the system iteratively updates its motion models and, as appropriate, further updates the coherent motion regions, and so forth. The system then does the same analysis for the remaining frames. The system next segments the image into regions of coherent motion and defines associated layers in terms of (i) pixel intensity values, (ii) associated motion model parameters, and (iii) order in "depth" within the image.
Inventors
- 👤 John Y. A. Wang
- 👤 Edward H. Adelson
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-10-28 17:44
CPC Classifications
H; H04; H04N; H04N19/00; H04N19/50; H04N19/503; H04N19/51; H04N19/537; H04N19/543; G; G06; G06T; G06T7/00; G06T7/20; G06T7/215; G06V; G06V10/00; G06V10/20; G06V10/28; H04N5/00; H04N5/14; H04N5/144; H04N5/145
Claims Analysis
Total Claims: 31
Claim 1
Independent Claim
286 words
A computer system for encoding digital image data relating to a sequence of image frames to represent the sequence of image frames as digital data organized as a plurality of ordered layers, the layers including information to be used to reproduce and display the same or an associated sequence of image frames, the system including: A. a local motion estimator, for producing local motion estimates for pixel neighborhoods in sequential frames and producing from these estimates associated dense motion models, each of the pixel neighborhoods including a selected pixel and the pixels adjacent to the selected pixel; B. a motion segmentation processor, for determining, jointly, from the dense motion models i. regions of coherent motion over the entire sequence of frames, and ii. for each region of coherent motion a motion model representing the transformation of the region from frame to frame for the sequence of frames; C. a layer extraction processor, for forming for each of the regions of coherent motion a layer by associating with the region data that represents the associated image data, the layer including i. data representing pixel intensity information relating to the sequence of frames for each of the pixels included in the region, ii. data representing an ordering of the layer relative to other layers by depth in the image, the layer extraction processor determining the depth of a layer by determining if pixels in the layer are occluded in any of the frames in the sequence; iii. data representing motion information including parameters relating to the motion model associated with the region by the motion segmentation processor, and D. means for outputting the digital data comprising the layers to reproduce the same or an associated sequence of frames.
Claim 2
Dependent Claim
286 words
A computer system for encoding digital image data relating to a sequence of image frames to represent the sequence of image frames as digital data organized as a plurality of ordered layers, the layers including information to be used to reproduce and display the same or an associated sequence of image frames, the system including: A. a local motion estimator, for producing local motion estimates for pixel neighborhoods in sequential frames and producing from these estimates associated dense motion models, each of the pixel neighborhoods including a selected pixel and the pixels adjacent to the selected pixel; B. a motion segmentation processor, for determining, jointly, from the dense motion models i. regions of coherent motion over the entire sequence of frames, and ii. for each region of coherent motion a motion model representing the transformation of the region from frame to frame for the sequence of frames; C. a layer extraction processor, for forming for each of the regions of coherent motion a layer by associating with the region data that represents the associated image data, the layer including i. data representing pixel intensity information relating to the sequence of frames for each of the pixels included in the region, ii. data representing an ordering of the layer relative to other layers by depth in the image, the layer extraction processor determining the depth of a layer by determining if pixels in the layer are occluded in any of the frames in the sequence; iii. data representing motion information including parameters relating to the motion model associated with the region by the motion segmentation processor, and D. means for outputting the digital data comprising the layers to reproduce the same or an associated sequence of frames.
Claim 3
Dependent Claim
79 words
The system of claim 1, wherein the motion segmentation processor includes: a. a motion estimator for producing estimates of parametric motion models for each of the regions of coherent motion; b. a clustering processor which groups similar motion models into clusters and produces for each cluster an estimate of a parametric motion model; c. a pixel assignment processor which associates individual pixels with the parametric motion model which most closely represents the local motion estimate associated with the pixel.
Claim 4
Dependent Claim
94 words
The system of claim 1, wherein the local motion estimator includes: a. a filter for filtering the image data to blur the images, the filtering producing filtered data; b. means for subsampling the filtered data to produce lower resolution versions of the images; c. means for determining estimates of small motions between selected pixel neighborhoods of the lower resolution versions of the images; and d. means for using the estimates of small motions in the lower resolution versions of the images to determine estimates of local motion in associated pixel neighborhoods of the image.
Claim 5
Dependent Claim
65 words
The system of claim 1, wherein the system further includes assembling means for assembling the layers produced by said layer extraction processor and producing a sequence of frames, said assembly means selectively assembling all the layers to reproduce the sequence of image frames or less than all of the layers to produce a sequence of frames in which selected regions of coherent motion are excluded.
Claim 6
Dependent Claim
74 words
The system of claim 4, wherein the system further includes: a. storage means for storing layers extracted from other sequences of image frames, b. said assembly means selectively assembling one or more of the layers stored in said storage means and one or more of the layers produced by said layer extraction processor to produce sequences of frames that include regions of coherent motion which are not present in the sequence of image frames.
Claim 7
Dependent Claim
46 words
The system of claim 4, wherein said system further includes: i. recognition means for identifying objects in the sequence of image frames, ii. said assembly means sending to said recognition means a sequence of frames with layers that represent objects that occlude other objects selectively excluded.
Claim 8
Dependent Claim
30 words
The system of claim 6, wherein said recognition means includes means for identifying an object by a predetermined characteristic in the image such as shape, intensity pattern, texture or color.
Claim 9
Dependent Claim
28 words
The system of claim 4, wherein said assembly means includes means for assembling the layers to produce frames which have wider fields of view than the image frames.
Claim 10
Dependent Claim
122 words
The system of claim 1, wherein the system further includes: a. interpolation means for interpolating motion parameters associated with the motion models determined by the motion segmentation processor, the interpolation means producing interpolated motion parameters representing inter-frame motion; b. frame generation means for generating frames which fall sequentially between the image frames, said frame generation means using the layers produced by the layer extraction processor and the motion parameters produced by the interpolation means to generate additional frames; and c. frame rate conversion means for producing a sequence of frames including the image frames and the additional frames, said frame rate conversion means producing a sequence of frames that has a frame rate which is different than the sequence of image frames.
Claim 11
Dependent Claim
77 words
The system of claim 9, wherein: i. the interpolation means includes means for determining interpolated pixel intensity values for portions of the image frame which are not included in the estimates produced by said local motion estimator, and ii. said system further includes frame generation means which uses the interpolation pixel intensity values produced by said interpolation means and the layers produced by said layer extraction processor to produce frames with higher resolution than the image frames.
Claim 12
Dependent Claim
29 words
The system of claim 10, wherein the frame generation means accumulates the interpolated pixel intensity values associated with selected frames and produces one or more frames with reduced noise.
Claim 13
Dependent Claim
51 words
The system of claim 9, wherein said system further includes: i. motion compensation means for manipulating selected frames of the sequence to produce motion-compensated pixel intensity information; ii. said frame generation means receiving the motion-compensated pixel intensity information and the layers and producing frames with higher resolution, reduced noise or both.
Claim 14
Dependent Claim
40 words
The system of claim 1, wherein said system further includes recognition means for receiving motion models and layers from said motion segmentation processor and said layer extraction processor, respectively, and identifying an object according to the motion of the object.
Claim 15
Independent Claim
200 words
A method of determining regions of coherent motion and associated motions in an image sequence which consists of a plurality of frames and using data associated with the regions of coherent motion to reproduce the same or an associated image sequence, the method comprising the steps of: A. estimating local motion of image pixels from frame to frame by determining motion estimates for selected pixels and pixel neighborhoods of the selected pixels; B. using the local motion estimates to determine iteratively regions of coherent motion in sequential frames and updating these regions of coherent motion for the plurality of frames; and C. for each region of coherent motion determining related motion parameters associated with frame to frame motion, based on the local motion estimates, and updating the motion parameters for updated regions of coherent motion; D. segmenting the image into regions of coherent motion by assigning each of the pixels to one or more regions of coherent motion associated with the motion models that most closely represent the local motions of the pixels; E. outputting data representing the image sequence as a plurality of regions of coherent motion to reproduce a display of the same or an associated image sequence.
Claim 16
Independent Claim
200 words
A method of determining regions of coherent motion and associated motions in an image sequence which consists of a plurality of frames and using data associated with the regions of coherent motion to reproduce the same or an associated image sequence, the method comprising the steps of: A. estimating local motion of image pixels from frame to frame by determining motion estimates for selected pixels and pixel neighborhoods of the selected pixels; B. using the local motion estimates to determine iteratively regions of coherent motion in sequential frames and updating these regions of coherent motion for the plurality of frames; and C. for each region of coherent motion determining related motion parameters associated with frame to frame motion, based on the local motion estimates, and updating the motion parameters for updated regions of coherent motion; D. segmenting the image into regions of coherent motion by assigning each of the pixels to one or more regions of coherent motion associated with the motion models that most closely represent the local motions of the pixels; E. outputting data representing the image sequence as a plurality of regions of coherent motion to reproduce a display of the same or an associated image sequence.
Claim 17
Dependent Claim
100 words
The method of determining regions of coherent motion and associated motions of claim 14, wherein the step of estimating local motion of the image pixels includes the steps of: a. filtering the image data to blur the images, the filtering producing filtered data; b. subsampling the filtered data to produce lower resolution versions of the images; c. determining estimates of small motions between selected pixel neighborhoods of the lower resolution versions of the images; and d. using the estimates of small motions in the lower resolution versions of the images to determine the estimates of local motion of the pixels.
Claim 18
Dependent Claim
102 words
The method of determining regions of coherent motion and associated motions of claim 14, wherein the step of segmenting the image into coherent motion regions includes the steps of: a. estimating parametric motion models for each of the regions; b. grouping similar parametric motion models into clusters and producing for each cluster an updated affine motion model; c. repeating step b a predetermined number of times or until no significant changes to the groupings are made; and d. associating individual pixels with the particular motion model produced in step b that most closely resembles the local motion estimate associated with the pixel.
Claim 19
Dependent Claim
94 words
The method of determining regions of coherent motion and associated motions of claim 16, wherein the step of segmenting the image into coherent motion regions further includes the steps of: e. iteratively updating the estimates of parametric motion models using regions determined by associating individual pixels with the motion models in step d; f. iteratively updating the clusters of models using the parametric motion models determined in step e; and g. repeating steps a-e until either a maximum number of iterations have been performed or the clusters do not change significantly between successive iterations.
Claim 20
Dependent Claim
34 words
The method of determining regions of coherent motion and associated motions of claim 14, wherein the step of segmenting the image into regions of coherent motion further includes segmenting the image into overlapping regions.
Claim 21
Dependent Claim
103 words
The method of determining regions of coherent motion and associated motions of claim 14 further including the steps of: F. assigning each of the regions of coherent motion to a separate layer; and G. determining for each of the pixels in each of the regions of coherent motion a pixel intensity across the plurality of frames and ordering the layers by depth in the image based on a comparison of the intensities of the pixels associated with the various regions; H. storing the image sequence as a series of ordered layers with each layer including pixel information, pixel intensities and related motion parameters.
Claim 22
Dependent Claim
24 words
The method of claim 19, wherein the method further includes the step of assembling the ordered layers to reproduce the sequence of image frames.
Claim 23
Dependent Claim
35 words
The method of claim 20, wherein the step of assembling the layers includes the step of assembling less than all the ordered layers to produce a sequence of frames in which selected layers are excluded.
Claim 24
Dependent Claim
65 words
The method of claim 20, wherein the method further includes the steps of: a. storing layers extracted from other sequences of image frames; and b. assembling one or more of the stored layers and one or more of the ordered layers to produce sequences of frames that include regions of coherent motion which are not present in the image sequence associated with the ordered layers.
Claim 25
Dependent Claim
44 words
The method of claim 20, wherein the method further includes: i. in the step of assembling layers, assembling layers to generate a frame in which objects that occlude other objects are selectively excluded; and ii. identifying an object in the generated frame by shape.
Claim 26
Dependent Claim
33 words
The method of claim 20, wherein the step of assembling the layers further includes assembling the layers to produce frames which have wider fields of view than the frames in the image sequence.
Claim 27
Dependent Claim
31 words
The method of claim 19, wherein said method further includes the step of analyzing the motion parameters and the layers and identifying an object according to the motion of the object.
Claim 28
Dependent Claim
90 words
The method of claim 14, wherein the method further includes the steps of: a. interpolating motion parameters to produce interpolated motion parameters relating to motion between frames; b. generating additional frames which fall sequentially between the frames in the image sequence using the regions of coherent motion and the interpolated motion parameters; and c. producing a sequence of frames including the frames of the image sequence and the additional frames to produce a sequence of frames that has a frame rate which is different than that of the image sequence.
Claim 29
Dependent Claim
70 words
The method of claim 26, wherein: i. the step of interpolating motion parameters further includes determining interpolated pixel intensity values for portions of the image for which local motion estimates have not been made; and ii. the step of generating frames further includes using the interpolated pixel intensity values produced in step i and the ordered layers to generate frames with higher resolution than the frames in the image sequence.
Claim 30
Dependent Claim
33 words
The method of claim 26, wherein the step of generating frames includes using the interpolated motion parameters to produce a sequence of high-resolution frames with a different frame rate than the image sequence.
Claim 31
Dependent Claim
55 words
The method of claim 26, wherein: i. said step of interpolating motion parameters further includes manipulating selected frames to produce motion-compensated pixel intensity information; and ii. said frame generating step further includes receiving the motion-compensated pixel intensity information and the ordered layers and generating one or more frames with higher resolution, reduced noise or both.
References Cited (Backward Citations)
US5214504A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 1999-09-17 | 2000-03-18 | 2000-09-17 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2003-09-17 | 2004-03-18 | 2004-09-17 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2007-09-17 | 2008-03-18 | 2008-09-17 | $7700.00 | Window Closed / Lapsed |