WO2010050911A1 - Method and system for image resizing based on interpolation enhanced seam operations - Google Patents
Method and system for image resizing based on interpolation enhanced seam operations Download PDFInfo
- Publication number
- WO2010050911A1 WO2010050911A1 PCT/US2008/012282 US2008012282W WO2010050911A1 WO 2010050911 A1 WO2010050911 A1 WO 2010050911A1 US 2008012282 W US2008012282 W US 2008012282W WO 2010050911 A1 WO2010050911 A1 WO 2010050911A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- seam
- image
- energy
- low
- seams
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 66
- 238000001514 detection method Methods 0.000 claims description 16
- 230000001172 regenerating effect Effects 0.000 claims description 2
- 230000007423 decrease Effects 0.000 claims 1
- 230000001747 exhibiting effect Effects 0.000 abstract 1
- 238000010586 diagram Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 7
- 230000000717 retained effect Effects 0.000 description 3
- 238000005303 weighing Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 2
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000003780 insertion Methods 0.000 description 1
- 230000037431 insertion Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000011946 reduction process Methods 0.000 description 1
- 238000001881 scanning electron acoustic microscopy Methods 0.000 description 1
- 238000004826 seaming Methods 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Classifications
-
- G06T3/04—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformation in the plane of the image
- G06T3/40—Scaling the whole image or part thereof
Definitions
- a seam is defined as an 8-connected curve extending vertically, from top to bottom, or extending horizontally, from the left side to the right side of the image.
- a pixel- wise energy map is pre-calculated, where energy is defined as the gradient intensity or as other energy measures.
- the lowest energy curves in the horizontal and/or vertical directions are determined by dynamic programming. To reduce the size of the image, the lowest energy curves are removed, as critical image contents are unlikely to be presented in low energy sections of the image. Enlarging the size of the images can be similarly performed by seam insertion.
- a system for resizing an image includes a low-energy seam detection module configured to define a low- energy seam in the image and to determine seams neighboring the low-energy seam, an interpolator configured to interpolate neighboring seams into a new seam and an image regenerator configured to combine the new seam to the image to resize the image.
- FIG. 1 depicts a high level flow diagram of a method for resizing an image by interpolating seams neighboring a low-energy seam in accordance with an embodiment of the present principles
- FIG. 2 depicts a pixel wise energy map illustrating a seam operation and interpolation for horizontal seams based on a one-dimensional manifold in accordance with an embodiment of the present principles
- FIG. 3 depicts a pixel wise energy map illustrating a seam operation and interpolation for a horizontal seam band in accordance with an embodiment of the present principles
- Embodiments of the present principles advantageously provide methods and systems for resizing an image by utilizing interpolation enhanced seam operations in accordance with various embodiments of the present invention.
- the present principles will be described primarily within the context of an image resizing system, the specific implementations of the present principles should not be treated as limiting the scope of the present principles. It will be appreciated by those skilled in the art and informed by the teachings of the present principles that the concepts of the present principles can be advantageously applied in other types of data representations. For example, the concepts of the present principles can be implemented in images included in video files.
- processor or “controller” should not be construed to refer exclusively to hardware capable of executing software, and can implicitly include, without limitation, digital signal processor (“DSP”) hardware, read-only memory (“ROM”) for storing software, random access memory (“RAM”), and non-volatile storage.
- DSP digital signal processor
- ROM read-only memory
- RAM random access memory
- FIGS. 1 and 2 an exemplary method 100 for resizing an image using interpolation enhanced seam operations in accordance with an embodiment of the present invention is illustrated. Aspects of the present principles are described herein with respect to an image represented by a square array in which each square corresponds to a pixel. However, it should be noted that other image portion sizes and configurations can be employed in various embodiments of the present principles. Thus, use of the term "pixel" herein incorporates other suitable image portion sizes and configurations.
- method 100 can be implemented to reduce the size of an image.
- the method 100 begins at step 102 in which a low-energy seam in an image is defined as discussed above. That is, a seam is defined as an 8-connected curve extending vertically, from top to bottom, or extending horizontally, from the left side to the right side of the image.
- a pixel-wise energy map is pre- calculated, where energy is defined as the gradient intensity or as other energy measures.
- the low-energy seam can comprise the seam with the lowest energy among the seams of the image being resized.
- the low-energy seam can be the lowest energy horizontal one-dimensional manifold on the image satisfying an 8-connection property.
- the low-energy seam is simply removed and pixels below it are shifted up by one pixel, resulting in abruptness and discontinuity.
- neighboring seams or pixels can be interpolated in accordance with one or more aspects of the present principles.
- the method can then proceed to step 104.
- seams neighboring the low-energy seam are determined.
- the low-energy seam determined in step 102 can be defined as a digitized curve according to equation one (1), which follows:
- Sv n are neighboring seams
- r n is the set of pixels on the neighboring seams
- (xnj, yn j ) are the coordinates of point r n j
- (X j , y j ) are the coordinates of pixels on the low-energy seam
- h is the height of the image in units of pixels. Similar to the horizontal seam example, two seams satisfy equation (4), where one neighboring seam is the set of pixels directly to the right of the low-energy seam pixels in the same rows as the low-energy seam pixels and another neighboring seam is a second set of pixels that is directly to the left of the low-energy seam pixels in the same rows as the low-energy seam pixels.
- the method 100 can then proceed to step 106.
- neighboring seams can be interpolated to form a new seam.
- interpolation methods in accordance with one or more embodiments of the present principles are based on one-dimensional manifolds in a two dimensional image plane. In this way, the interpolation can be based on a grid defined by a seam curve.
- the neighboring seams can be interpolated to form a new seam in accordance with equation five (5), which follows:
- ⁇ k is a weighing factor for each pixel t n ⁇ ⁇ neighboring the horizontal low-energy seam
- tj' is the set of pixels corresponding to the new seam.
- the weighing factor ⁇ k can be defined as a function of a neighboring pixel's energy e k and its distance d k to the pixel tj on the low-energy seam corresponding to tj' according to equation six (6), which follows:
- f() is a monotonically increasing function of ⁇ j and is a monotonically decreasing function of d k .
- a simple example of the function, f() can be characterized according to equation seven (7), which follows:
- neighboring seams can be interpolated to form a new seam in accordance with equation eight (8), which follows:
- ⁇ k is a weighing factor, described above, for each pixel r nk neighboring the vertical low energy seam and ⁇ ' is the set of pixels corresponding to the new seam.
- the neighboring seam with the lowest energy among the neighboring seams is removed. For example, removal of a horizontal neighboring seam with the lowest energy can reduce the image height to h-1. Similarly, removal of a vertical neighboring seam with the lowest energy can reduce the image width to w-1 , in units of pixels.
- the energy of the two neighboring seams can be calculated from a pre-computed energy map. Rather than simply removing the low-energy seam, a continuous and smoother image is retained due to an amalgamated new seam in accordance with the above described embodiment of the present invention. The method can then optionally proceed to step 1 12.
- step 112 it is determined whether the image is the desired size. If the image is the desired size, then the method 100 can be exited. However, if the image is not the desired size, the method can proceed to step 114. At step 114, the energy of the new seam can be calculated for further processing to reconfigure the picture to a desired size. That is, thereafter, steps 102-112 can be repeated.
- steps 102- 106 can be performed as described above.
- the method 100 can then proceed to step 108.
- the new seam formed from interpolation is combined with the image.
- the new seam can be added in a position adjacent to the low-energy seam to increase the image height to h+1 if horizontal resizing is performed, or to increase the image width to w+1 if vertical resizing is performed.
- step 110 can be skipped and steps 112 and 114 can be performed as described above.
- the low-energy seam can be a one dimensional manifold, in either the horizontal or vertical direction with the lowest energy in an image. Because of the continuity of the image, the approximate area around a seam with a width of one pixel has a high probability of having low energy.
- a seam band can be removed or added to expedite the resizing process while retaining continuity and smoothness of an image by utilizing interpolation methods in accordance with the present invention and as described herein. Referring again to FIG. 1 with continuing reference to FIG. 3, an image reduction process utilizing seam bands is described herein.
- a low- energy seam can be defined.
- the low-energy seam comprises a seam band with a width of N+1 for horizontal seam bands or M+1 for vertical seam bands in units of pixels.
- the width of the seam band is determined in accordance with a desired quality of a resized image.
- the central seam of the seam band can correspond to the seam with the lowest energy among the seams of the image being resized.
- the central seam can be defined by equation (1 ) for horizontal seams and can be defined by equation (3) for vertical seams, described above.
- a horizontal seam band can be defined, in addition to the central seam, by pixels satisfying the equation nine (9), which follows:
- t B is the set of pixels on the seam band in addition to the central seam
- (x B i, y ⁇ i) are the coordinates of point t B j
- (Xi, yi) are coordinates of pixels on the central seam
- w is the width of the image in units of pixels
- the width of the seam band is N+1 in units of pixels.
- a vertical seam band can be defined, in addition to the central seam, by pixels satisfying the equation ten (10), which follows:
- r B is the set of pixels on the seam band in addition to the central seam
- (x Bj , y Bj ) are the coordinates of point r B ⁇
- (X j , y,) are coordinates of pixels on the central seam
- h is the height of the image in units of pixels
- the width of the seam band is M+1 in units of pixels.
- seam band 302 in FIG. 3 is one example of a horizontal seam band.
- N is 2.
- seams neighboring the low-energy seam are determined.
- the seams neighboring the low-energy seam can be defined by equation eleven (11), which follows:
- BVn ⁇ r ⁇ nj : (xBnj, yBnj)
- + 1,
- 0 > , (12)
- B Vn are neighboring seams
- r Bn is the set of pixels on the neighboring seams
- (XB ⁇ J, y ⁇ n j ) are the coordinates of point r Bnj
- (x j , Yj) are the coordinates of pixels on the low-energy seam
- h is the height of the image in units of pixels
- the width of the low-energy seam is M+1.
- two seams satisfy equation (12), where one neighboring seam is the set of pixels directly to the right of the low-energy seam pixels in the same rows as the low-energy seam pixels and a second neighboring seam is a set of pixels that is directly to the left of the low- energy seam pixels in the same rows as the low-energy seam pixels.
- the low-energy seam is a seam band with a width greater than one pixel.
- replacing the seam band with the new seam reduces the image size to a height of h-2 pixels for horizontal seam operations and a width of w-2 pixels for vertical seam operations.
- Steps 112 and 114 can be performed as discussed above.
- the method can be repeated until a desired image size is attained.
- the described method of the present invention can also be employed to increase the size of an image using a low-energy seam that comprises a seam band.
- steps 102-106 can be performed as described above.
- the method can proceed to step 108.
- the new seam formed from interpolation is combined with the image.
- the new seam can replace the low-energy seam.
- a new horizontal seam of a single pixel width can be multiplied to form a horizontal seam band of width N+1.
- a new vertical seam of a single pixel width can be multiplied to form a seam band of width M+1 for vertical seam bands.
- new horizontal seam 308 can be multiplied to a width of 3 to form horizontal seam 310, which includes 3 parallel and equal seams 310-1 , 308, and 310-2.
- the width of seam 310 can be increased by adding a number of parallel seams equal to the new seam 308 between neighboring seams 304 and 308.
- the width of the new seam can be adjusted to a width of N+a to increase the image height to h+(a-1 ) if horizontal resizing is performed.
- the width of the new seam can be adjusted to a width of M+b to increase the image width of w+(b-1) if vertical resizing is performed.
- step 110 can be skipped and steps 112 and 114 can be carried out as described above.
- interpolation can be performed after employing seam operations to resize an image.
- the duplicate pixels can be added directly above or below pixels in the low-energy seam in the same columns as pixels in the low-energy seam. If the low-energy seam is a vertical seam, then the duplicate pixels can be added directly to the right or the left of pixels in the low-energy seam in the same rows as pixels in the low- energy seam. Moreover, more than one duplicate can be added to the image to increase its image size. The method then proceeds to step 408.
- each pixel of each neighborhood is regenerated by interpolating their adjacent pixels.
- binary maps can be generated to represent pixels in a neighborhood surrounding each pixel in a low- energy seam.
- the neighborhood pixels can be flagged as "1" and non- neighborhood pixels are flagged as "0".
- MAP v (l,m) and MAP h (l,m) represent the vertical and horizontal binary maps, respectively.
- the regenerated pixels replace their corresponding pixels in the content-aware resized image to form a smooth and continuous image. The method proceeds to step 410.
- the energy of the regenerated pixels can be calculated for further processing to configure the image to a desired size. That is, thereafter, steps 402- 410 can be repeated.
- steps 402- 410 can be repeated.
- both horizontal and vertical resizing can be performed.
- one suitable configuration is to alternate the performance of the above described method, method 400, of the present invention between horizontal and vertical seam and the performance of the interpolation operations until a desired size is reached.
- the system 500 of FIG. 5 illustratively includes a low-energy seam detection module 502, an energy map generator 504, an interpolator 506, and an image regenerator 508.
- the system 500 can be utilized in one or more processors running software.
- each system element can be a software component run on one or more processors with corresponding memory.
- an input 502 comprising an image can be provided to the low-energy seam detection module 504.
- the low-energy seam detection module 504 can determine a low-energy seam of an image, as discussed above with respect to steps 102 and 402. To aid in defining a low-energy seam, the low-energy seam detection module 504 can utilize an energy map generator 506 to generate one or more energy maps of the image. Further, the low-energy seam detection module 504 can determine seams neighboring the low- energy seam, as discussed above with respect to step 104, and/or can determine neighborhoods of pixels surrounding the low-energy seam, as discussed above with respect to step 404. Thereafter, the energy values and the pixel brightness information for pixels in the low-energy seam, the neighboring seams and/or neighborhoods of pixels surrounding the low-energy seam can be provided to the interpolator 508.
- the low-energy seam detection module 504 can provide, to the image regenerator 510, an indication of where the low-energy seam is located in the image as well as an indication of the location of neighboring seams in the image and/or the locations of neighborhoods of pixels in the image with their corresponding energy values
- the interpolator 508 can interpolate neighboring seams to form a new seam, as discussed above with respect to step 106. Moreover, the interpolator 508 can interpolate pixels adjacent to pixels in the neighborhoods surrounding pixels in the low-energy seam, as discussed above with respect to step 408. It should be noted that pixel brightness information and pixel distances can be received from the low-energy seam detection module 504. Alternatively, pixel brightness information and pixel distances can be determined by the interpolator 508 from the image of input 502, which can be received from the low-energy seam detection module 504, for example. Subsequent to interpolation, the pixel brightness information for the new seam and/or for the neighborhoods surrounding pixels in the low-energy seam can be provided to the image regenerator 510.
- the image regenerator 510 can receive the image 502. As stated above, the image regenerator 510 can also receive the locations of the low-energy seam, the neighboring seams, and/or the neighborhoods of pixels surrounding the low-energy seam with their corresponding energy values from the low-energy seam detection module 504. Further, as stated above, pixel brightness information for the new seam and/or for the neighborhoods of pixels surrounding pixels in the low-energy seam can be received from the interpolator 508. Utilizing data received from the low-energy seam detection module 504 and from the interpolator 508, the image regenerator adds a new seam to the image and optionally removes the neighboring seam with the lowest energy, as discussed above with respect to steps 108 and 1 10. In addition, the image regenerator 510 can resize the image and regenerate pixels in the neighborhoods surrounding the pixels in the low-energy seam, as discussed above with respect to steps 406 and 408.
- the low-energy seam detection module 504 can calculate the energy of the new seam and/or regenerated pixels, as discussed above with respect to steps 114 and/or 412, utilizing the energy map generator 506.
Abstract
Description
Claims
Priority Applications (7)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN200880131822.6A CN102203825B (en) | 2008-10-29 | The method and system adjusted for image size based on interpolation enhanced seaming operation | |
EP08876357.8A EP2347385B1 (en) | 2008-10-29 | 2008-10-29 | Method and system for image resizing based on interpolation enhanced seam operations |
BRPI0823189-3A BRPI0823189A2 (en) | 2008-10-29 | 2008-10-29 | Method and system for resizing images based on tween-optimized sewing operations |
JP2011534458A JP5451767B2 (en) | 2008-10-29 | 2008-10-29 | Image resizing method and system based on interpolation enhanced seam operation |
KR1020117012144A KR101617059B1 (en) | 2008-10-29 | 2008-10-29 | Method and system for image resizing based on interpolation enhanced seam operations |
US12/998,387 US9741092B2 (en) | 2008-10-29 | 2008-10-29 | Method and system for image resizing based on interpolation enhanced seam operations |
PCT/US2008/012282 WO2010050911A1 (en) | 2008-10-29 | 2008-10-29 | Method and system for image resizing based on interpolation enhanced seam operations |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/US2008/012282 WO2010050911A1 (en) | 2008-10-29 | 2008-10-29 | Method and system for image resizing based on interpolation enhanced seam operations |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2010050911A1 true WO2010050911A1 (en) | 2010-05-06 |
Family
ID=40886756
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/US2008/012282 WO2010050911A1 (en) | 2008-10-29 | 2008-10-29 | Method and system for image resizing based on interpolation enhanced seam operations |
Country Status (6)
Country | Link |
---|---|
US (1) | US9741092B2 (en) |
EP (1) | EP2347385B1 (en) |
JP (1) | JP5451767B2 (en) |
KR (1) | KR101617059B1 (en) |
BR (1) | BRPI0823189A2 (en) |
WO (1) | WO2010050911A1 (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2013084202A (en) * | 2011-10-12 | 2013-05-09 | Nippon Telegr & Teleph Corp <Ntt> | Image processing method, image processing device, and image processing program |
JP2013531308A (en) * | 2010-07-09 | 2013-08-01 | イー・エヌ・エール・イー・アー−アンスティチュ・ナシオナル・ドゥ・ラ・ルシェルシュ・アン・ナンフォルマティーク・エ・タン・ノトマティーク | Image synthesizer |
Families Citing this family (19)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8280191B1 (en) | 2008-07-31 | 2012-10-02 | Abode Systems Incorporated | Banded seam carving of images with pyramidal retargeting |
US8265424B1 (en) | 2008-07-31 | 2012-09-11 | Adobe Systems Incorporated | Variable seam replication in images with energy-weighted priority |
US8290300B2 (en) * | 2008-07-31 | 2012-10-16 | Adobe Systems Incorporated | Seam-based reduction and expansion of images with color-weighted priority |
US8270765B1 (en) | 2008-07-31 | 2012-09-18 | Adobe Systems Incorporated | Hybrid seam carving and scaling of images with configurable energy threshold |
US8280186B1 (en) | 2008-07-31 | 2012-10-02 | Adobe Systems Incorporated | Seam-based reduction and expansion of images with table-based priority |
US8280187B1 (en) | 2008-07-31 | 2012-10-02 | Adobe Systems Incorporated | Seam carving and expansion of images with color frequency priority |
US8270766B1 (en) | 2008-07-31 | 2012-09-18 | Adobe Systems Incorporated | Hybrid seam carving and scaling of images with configurable carving tolerance |
US8625932B2 (en) * | 2008-08-28 | 2014-01-07 | Adobe Systems Incorporated | Seam carving using seam energy re-computation in seam neighborhood |
US8581937B2 (en) | 2008-10-14 | 2013-11-12 | Adobe Systems Incorporated | Seam-based reduction and expansion of images using partial solution matrix dependent on dynamic programming access pattern |
US8270771B2 (en) * | 2008-12-09 | 2012-09-18 | Xerox Corporation | Iterative selection of pixel paths for content aware image resizing |
US8963960B2 (en) | 2009-05-20 | 2015-02-24 | Adobe Systems Incorporated | System and method for content aware hybrid cropping and seam carving of images |
US8358876B1 (en) | 2009-05-20 | 2013-01-22 | Adobe Systems Incorporated | System and method for content aware in place translations in images |
US8659622B2 (en) | 2009-08-31 | 2014-02-25 | Adobe Systems Incorporated | Systems and methods for creating and editing seam carving masks |
US8213745B2 (en) * | 2009-10-09 | 2012-07-03 | Eastman Kodak Company | Seam carving for image resizing |
US8184928B2 (en) * | 2009-10-20 | 2012-05-22 | Eastman Kodak Company | Combining seam carving an image resizing |
JP5719271B2 (en) * | 2011-10-12 | 2015-05-13 | 日本電信電話株式会社 | Image processing method, image processing apparatus, and image processing program |
US8873887B2 (en) * | 2013-01-24 | 2014-10-28 | Google Inc. | Systems and methods for resizing an image |
US11272209B2 (en) | 2018-04-03 | 2022-03-08 | Samsung Electronics Co., Ltd. | Methods and apparatus for determining adjustment parameter during encoding of spherical multimedia content |
US20220020113A1 (en) * | 2020-07-15 | 2022-01-20 | Instasize, Inc. | Image resizing using seam carving |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1968008A2 (en) * | 2007-03-06 | 2008-09-10 | Mitsubishi Electric Corporation | Method for content-aware image retargeting |
Family Cites Families (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
GB9504307D0 (en) | 1995-03-03 | 1995-04-19 | Philips Electronics Uk Ltd | Video image processing |
US7839422B2 (en) | 2006-12-13 | 2010-11-23 | Adobe Systems Incorporated | Gradient-domain compositing |
US8400473B2 (en) * | 2009-06-24 | 2013-03-19 | Ariel Shamir | Multi-operator media retargeting |
EP2460140B1 (en) * | 2009-07-30 | 2013-09-11 | TP Vision Holding B.V. | Distributed image retargeting |
US8373802B1 (en) * | 2009-09-01 | 2013-02-12 | Disney Enterprises, Inc. | Art-directable retargeting for streaming video |
US8213745B2 (en) * | 2009-10-09 | 2012-07-03 | Eastman Kodak Company | Seam carving for image resizing |
US8184928B2 (en) * | 2009-10-20 | 2012-05-22 | Eastman Kodak Company | Combining seam carving an image resizing |
US8494302B2 (en) * | 2010-11-11 | 2013-07-23 | Seiko Epson Corporation | Importance filtering for image retargeting |
-
2008
- 2008-10-29 EP EP08876357.8A patent/EP2347385B1/en active Active
- 2008-10-29 BR BRPI0823189-3A patent/BRPI0823189A2/en not_active Application Discontinuation
- 2008-10-29 US US12/998,387 patent/US9741092B2/en active Active
- 2008-10-29 WO PCT/US2008/012282 patent/WO2010050911A1/en active Application Filing
- 2008-10-29 JP JP2011534458A patent/JP5451767B2/en active Active
- 2008-10-29 KR KR1020117012144A patent/KR101617059B1/en active IP Right Grant
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1968008A2 (en) * | 2007-03-06 | 2008-09-10 | Mitsubishi Electric Corporation | Method for content-aware image retargeting |
Non-Patent Citations (1)
Title |
---|
AVIDAN S ET AL: "Seam carving for content-aware image resizing", ACM TRANSACTIONS ON GRAPHICS, ACM, US, vol. 26, no. 3, 1 July 2007 (2007-07-01), pages 10 - 1, XP007904203, ISSN: 0730-0301 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2013531308A (en) * | 2010-07-09 | 2013-08-01 | イー・エヌ・エール・イー・アー−アンスティチュ・ナシオナル・ドゥ・ラ・ルシェルシュ・アン・ナンフォルマティーク・エ・タン・ノトマティーク | Image synthesizer |
JP2013084202A (en) * | 2011-10-12 | 2013-05-09 | Nippon Telegr & Teleph Corp <Ntt> | Image processing method, image processing device, and image processing program |
Also Published As
Publication number | Publication date |
---|---|
EP2347385A1 (en) | 2011-07-27 |
JP2012507923A (en) | 2012-03-29 |
JP5451767B2 (en) | 2014-03-26 |
KR20110091700A (en) | 2011-08-12 |
EP2347385B1 (en) | 2020-03-04 |
BRPI0823189A2 (en) | 2015-06-23 |
US20110200274A1 (en) | 2011-08-18 |
KR101617059B1 (en) | 2016-04-29 |
US9741092B2 (en) | 2017-08-22 |
CN102203825A (en) | 2011-09-28 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
EP2347385B1 (en) | Method and system for image resizing based on interpolation enhanced seam operations | |
US9665542B2 (en) | Determining median value of an array on vector SIMD architectures | |
US7149355B2 (en) | Image processing apparatus, image processing method, image processing program, and computer-readable record medium storing image processing program | |
RU2623806C1 (en) | Method and device of processing stereo images | |
CN110246081B (en) | Image splicing method and device and readable storage medium | |
CN108346131A (en) | A kind of digital image scaling method, device and display equipment | |
CN103390267A (en) | Image processing method and device | |
US7825928B2 (en) | Image processing device and image processing method for rendering three-dimensional objects | |
US8295647B2 (en) | Compressibility-aware media retargeting with structure preserving | |
JPH1097644A (en) | Method for expressing object direction image using irregular mesh and its device | |
CN113506305B (en) | Image enhancement method, semantic segmentation method and device for three-dimensional point cloud data | |
CN114648458A (en) | Fisheye image correction method and device, electronic equipment and storage medium | |
JP2010515131A (en) | Method and system for generating boundaries in the process of rasterizing vector graphics, and method for manufacturing the system | |
JP2022153857A (en) | Image processing apparatus, image processing method, moving device, and computer program | |
US20200019803A1 (en) | Method and apparatus for determining summation of pixel characteristics for rectangular region of digital image avoiding non-aligned loads using multiple copies of input data | |
CN115526903A (en) | Hardware computing system and method for image upsampling based on Canny algorithm | |
WO2011121563A1 (en) | Detecting saliency in an image | |
JP4636526B2 (en) | Method for correcting non-functional pixels in digital X-ray imaging in real time | |
US20170270632A1 (en) | Caching Method Of Graphic Processing Unit | |
GB2605360A (en) | Method, Apparatus and Storage Medium for Realizing Geometric Viewing Frustum of OCC Tree in Smart City | |
KR20150019192A (en) | Apparatus and method for composition image for avm system | |
RU168781U1 (en) | STEREO IMAGE PROCESSING DEVICE | |
KR20170109465A (en) | Apparatus for reconstructing image and method using the same | |
CN104978712A (en) | Image processing method and device | |
CN110807113A (en) | Non-iterative elimination method for rectangular primitive overlap in visual layout |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 200880131822.6 Country of ref document: CN |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 08876357 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 12998387 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2011534458 Country of ref document: JP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2008876357 Country of ref document: EP |
|
ENP | Entry into the national phase |
Ref document number: 20117012144 Country of ref document: KR Kind code of ref document: A |
|
ENP | Entry into the national phase |
Ref document number: PI0823189 Country of ref document: BR Kind code of ref document: A2 Effective date: 20110426 |