Digital image manipulation
View Patent ↗There is disclosed a method for seamlessly replacing areas in a digital image with corresponding data from temporally close digital images depicting substantially the same scene. The method uses localized image registration error minimization over a fixed preliminary boundary. A least cost closed path which constitutes a boundary for the area to be replaced is calculated using dynamic programming. The replacement area is blended such that image data information from one image is seamlessly replaced with image data information from another image.
1. A computerized method for digital image manipulation comprising:
receiving a source image and a target image;
identifying a source area in the source image, the source area having a set of coordinates;
using the set of coordinates for the source area to, in response to said identifying said source area, identify a target area in the target image corresponding to the set of coordinates of the source area, wherein identifying said target area further comprises finding a compensation movement between at least a part of said source image and at least a part of said target image, such that said compensation movement minimizes a difference between the source image and the target image; and
forming by a processor a digital image based on image data information of the target image, wherein image data information of the target area is seamlessly replaced with image data information of the source area.
2. A computerized method for digital image manipulation comprising:
receiving a source image and a target image;
identifying a source area in the source image, the source area having a set of coordinates;
using the set of coordinates for the source area to, in response to said identifying said source area, identify a target area in the target image corresponding to the set of coordinates of the source area, wherein identifying said target area further comprises finding a compensation movement between at least a part of said source image and at least a part of said target image, such that said compensation movement minimizes a difference between the source image and the target image; and
forming by a processor a digital image based on image data information of the source image, wherein image data information of the source area is seamlessly replaced with image data information of the target area.
3. The method according to claim 1 or 2 , further comprising
receiving feature information relating to a feature of at least one of said source image and said target image, wherein at least one of said source area and said target area comprises said feature; and
identifying said source area based on said feature.
4. The method according to claim 3 , further comprising
determining whether or not one of said source area and said target area fulfills a condition relating to said feature information; and
forming said digital image based on said determining.
5. The method according to claim 4 , wherein said condition relates to feature recognition of at least one from a group of human person, human face, human mouth, human eye.
6. The method according to claim 1 or 2 , wherein the source image is selected from a plurality of possible source images.
7. The method according to claim 6 , wherein each one of said plurality of possible source images is associated with a distortion measure, and wherein the source image is selected from the plurality of possible source image as the image having smallest distortion measure.
8. The method according to claim 1 or 2 , wherein the source image and the target image are received within a predetermined time interval.
9. The method according to claim 1 or 2 , wherein the source image and the target are associated with different image capturing features.
10. The method according to claim 9 , wherein said image capturing features relate to at least one from exposure, resolution, focus, blur, and flash level.
11. The method according to claim 1 or 2 , identifying the source area comprises selecting an area or object with a man machine interface.
12. The method according to claim 1 or 2 , wherein image data information of the source image and image data information of the target image represent substantially the same scene.
13. The method according to claim 1 or 2 , wherein said difference is determined for at least a part of the target image and relates to at least one from a translation, rotation, or projection of at least part of the source image in relation to said target image.
14. The method according to claim 1 or 2 , further comprising using a thin plate spine method to find said compensation movement.
15. The method according to claim 1 or 2 , further comprising determining a detail measure between a candidate source area in said source image and a candidate target area in said target image, said source area and said target area being identified based on the determined detail measure.
16. The method according to claim 15 , further comprising identifying the source area and the target area by calculating a cut using the candidate source area and the candidate target area, said cut defining a border enclosing the source area, and said cut being determined such that said detail measure is minimized.
17. The method according to claim 15 , wherein said detail measure is based on data unit lengths of data units in said candidate source area and said candidate target area.
18. The method according to claim 1 or 2 , further comprising blending the source image and the target image using quad-tree optimization.
19. The method according to claim 1 or 2 , further comprising replacing one of the target image and the source image with the formed digital image.
20. The method according to claim 1 or 2 , wherein the source image and the target image have been encoded, said method further comprising in the source image decoding only image data information representing the source area; and in the target image decoding only image data information representing the target area.
21. The method according to claim 1 or 2 , wherein the source area corresponds to a source object, wherein the target area corresponds to a target object, and wherein the source object and the target object represent one and the same object.
22. The method according to claim 1 or 2 , wherein the source image and the target image originate from a video sequence.
23. The method according to claim 1 or 2 , further comprising receiving a signal identifying said source area.
24. A non-transitory computer-readable medium comprising instructions that when downloaded to a computer is configured to perform a method according to claim 1 or 2 .
25. The method according to claim 1 or 2 , wherein image data information of the source image and image data information of the target image represent substantially the same scene.
26. An apparatus for digital image manipulation comprising:
a processor configured to receive a source image and a target image;
an interface configured to identify a source area in the source image, the source area having a set of coordinates;
said processor is further configured to use the set of coordinates for the source area to, in response to said identifying said source area, identify a target area in the target image corresponding to the set of coordinates of the source area, wherein identifying said target area further comprises finding a compensation movement between at least a part of said source image and at least a part of said target image, such that said compensation movement minimizes a difference between the source image and the target image; and
form a digital image based on image data information of the target image, wherein image data information of the target area is seamlessly replaced with image data information of the source area.
27. An apparatus according to claim 26 , wherein the interface comprises a display and the display is configured to identify user input that identifies the source area.
28. An apparatus according to claim 27 , wherein the display is configured to provide a geometrical shape on the display in response to the user input.
29. An apparatus for digital image manipulation comprising:
a processor configured to receive a source image and a target image;
an interface configured to identify a source area in the source image, the source area having a set of coordinates;
said processor is further configured to use the set of coordinates for the source area to, in response to said identifying said source area, identify a target area in the target image corresponding to the set of coordinates of the source area, wherein identifying said target area further comprises finding a compensation movement between at least a part of said source image and at least a part of said target image, such that said compensation movement minimizes a difference between the source image and the target image; and
form a digital image based on image data information
of the source image, wherein image data information of the source area is
seamlessly replaced with image data information of the target area.
30. The Apparatus according to claim 26 or 29 , further comprising a camera and wherein said source image and said target image are received from said camera.