IPC분류정보
국가/구분 |
United States(US) Patent
등록
|
국제특허분류(IPC7판) |
|
출원번호 |
US-0957322
(2010-11-30)
|
등록번호 |
US-8803918
(2014-08-12)
|
발명자
/ 주소 |
- Georgiev, Todor G.
- Chunev, Georgi N.
|
출원인 / 주소 |
- Adobe Systems Incorporated
|
대리인 / 주소 |
|
인용정보 |
피인용 횟수 :
6 인용 특허 :
89 |
초록
▼
Methods, apparatus, and computer-readable storage media for calibrating focused plenoptic camera data. A calibration technique that does not modify the image data may be applied to raw plenoptic images. Calibration parameters, including but not limited to tilt angle, corner crops, main lens distance
Methods, apparatus, and computer-readable storage media for calibrating focused plenoptic camera data. A calibration technique that does not modify the image data may be applied to raw plenoptic images. Calibration parameters, including but not limited to tilt angle, corner crops, main lens distance from the microlens array, sensor distance from the microlens array, and microimage size, may be specified. Calibration may include scaling down the input texture coordinates for the plenoptic image so that the new coordinate range fits the size of the texture with crops taken into account. These coordinates may be further transformed by one or more of a matrix performing a scaling, to correct for lens distortion; a rotation, to correct for tilts; and a translation that finalizes the necessary corner crops. A transformation matrix is generated that can be applied to the raw image by radiance processing techniques such as super-resolution techniques.
대표청구항
▼
1. A method, comprising: obtaining, by a computing device, a radiance image comprising a plurality of separate microimages of an image of a scene, the radiance image having a spatial resolution that is a function of a resolution of said microimages and an amount of overlap in rendering of said micro
1. A method, comprising: obtaining, by a computing device, a radiance image comprising a plurality of separate microimages of an image of a scene, the radiance image having a spatial resolution that is a function of a resolution of said microimages and an amount of overlap in rendering of said microimages;displaying first and second portions of a user interface, the first portion comprising the radiance image, the second portion comprising a control area including multiple user interface elements that are user-selectable to adjust respective parameters of the radiance image;applying, by the computing device, a calibration technique to the radiance image to generate a transformation matrix for the radiance image without modifying the radiance image, the calibration technique being based on an adjusted value of one or more of the multiple user interface elements; andperforming, by the computing device, a radiance processing technique on the radiance image to render an output image, at least a portion of the radiance processing technique being performed by a fragment shades on one or more graphics processing units (GPUs), the radiance processing technique including at least: applying the transformation matrix to the radiance image to transform the plurality of separate microimages according to the transformation matrix before texture look-ups in the fragment shader; andrendering the output image from the transformed microimages. 2. The method as recited in claim 1, wherein the radiance processing technique is a super-resolution rendering technique, and wherein the output image is a super-resolved output image of the scene. 3. The method as recited in claim 1, wherein said transforming the plurality of separate microimages according to the transformation matrix comprises aligning one or more of the microimages according to one or more alignment parameters indicated by the transformation matrix. 4. The method as recited in claim 1, wherein said transforming the plurality of separate microimages according to the transformation matrix comprises scaling one or more of the microimages according to one or more scale parameters indicated by the transformation matrix. 5. The method as recited in claim 1, wherein said transforming the plurality of separate microimages according to the transformation matrix comprises cropping one or more of the microimages according to one or more crop parameters indicated by the transformation matrix. 6. The method as recited in claim 1, wherein said applying a calibration technique to the radiance image to generate a transformation matrix for the radiance image comprises: receiving alignment input via the user interface specifying one or more alignment parameters for microimages in the radiance image;receiving scaling input via the user interface specifying one or more scale parameters for microimages in the radiance image;receiving cropping input via the user interface specifying one or more crop parameters for microimages in the radiance image; andgenerating the transformation matrix for the radiance image according to the alignment parameters, scale parameters, and crop parameters. 7. The method as recited in claim 1, further comprising allowing the fragment shader to assume calibrated input images and avoid performing an operation on the radiance image that results in resampling or anti-aliasing the radiance image. 8. A system, comprising one or more processors; anda memory comprising program instructions, wherein the program instructions are executable by the at least one processor to: obtain a radiance image comprising a plurality of separate microimages of an image of a scene, the radiance image having a spatial resolution that is a function of a resolution of said microimages and an amount of overlap in rendering of said microimages, and not of a number of microlenses used to capture said microimages;display the radiance image in a first portion of a user interface and a control area in a second portion of the user interface, the control area including one or more user interface elements that are user-selectable to adjust respective parameters of the radiance image;apply a calibration technique to the radiance image to generate a transformation matrix for the radiance image without modifying the radiance image, the calibration technique being based on an adjusted value of the one or more user interface elements; andperform a radiance processing technique on the radiance image to render an output image, at least a portion of the radiance processing technique being performed by a fragment shader on one or more graphics processing units (GPUs), the radiance processing technique including at least: application of the transformation matrix to the radiance image, prior to texture lookups in the fragment shader, to transform the plurality of separate microimages according to the transformation matrix; anda rendering of the output image from the transformed microimages. 9. The system as recited in claim 8, wherein the radiance processing technique is a super-resolution rendering technique, and wherein the output image is a super-resolved output image of the scene. 10. The system as recited in claim 8, wherein at least one of the one or more processors is a graphics processing unit (GPU), and wherein at least a portion of the radiance processing technique is performed by program instructions executing on the at least one GPU. 11. The system as recited in claim 8, wherein, to transform the plurality of separate microimages according to the transformation matrix, the program instructions are executable by the at least one processor to align one or more of the microimages according to one or more alignment parameters indicated by the transformation matrix. 12. The system as recited in claim 8, wherein, to transform the plurality of separate microimages according to the transformation matrix, the program instructions are executable by the at least one processor to scale one or more of the microimages according to one or more scale parameters indicated by the transformation matrix. 13. The system as recited in claim 8, wherein, to transform the plurality of separate microimages according to the transformation matrix, the program instructions are executable by the at least one processor to crop one or more of the microimages according to one or more crop parameters indicated by the transformation matrix. 14. The system as recited in claim 8, wherein the system further comprises a display device, and wherein, to apply a calibration technique to the radiance image to generate a transformation matrix for the radiance image, the program instructions are executable by the at least one processor to: receive alignment input via the user interface specifying one or more alignment parameters for microimages in the radiance image;receive scaling input via the user interface specifying one or more scale parameters for microimages in the radiance image;receive cropping input via the user interface specifying one or more crop parameters for microimages in the radiance image; andgenerate the transformation matrix for the radiance image according to the alignment parameters, scale parameters, and crop parameters. 15. A computer-readable storage memory storing program instructions, wherein the program instructions are computer-executable to implement: obtaining a radiance image comprising a plurality of separate microimages of an image of a scene, the radiance image having a spatial resolution that is a function of a resolution of said microimages and an amount of overlap in rendering of said microimages;displaying, via a user interface having at least a first portion and a second portion, the radiance image in the first portion and a control area in the second portion, the control area including multiple user interface elements that are user-selectable to adjust respective parameters of the radiance image;applying a calibration technique to the radiance image to generate a transformation matrix for the radiance image without modifying the radiance image, the calibration technique being based on an adjusted value of one or more of the multiple user interface elements; andperforming a radiance processing technique on the radiance image to render an output image, at least a portion of the radiance processing technique being performed by a fragment shader on one or more graphics processing units (GPUs), the radiance processing technique being performed by at least: applying the transformation matrix to the radiance image to transform the plurality of separate microimages according to the transformation matrix, the transformation matrix being applied to the radiance image before texture look-ups in the fragment shader to allow shaders to be generated that assume calibrated images; andrendering the output image from the transformed microimages. 16. The computer-readable storage memory as recited in claim 15, wherein the radiance processing technique is a super-resolution rendering technique, and wherein the output image is a super-resolved output image of the scene. 17. The computer-readable storage memory as recited in claim 15, wherein the program instructions are computer-executable to implement performing at least a portion of the radiance processing technique on one or more graphics processing units (GPUs). 18. The computer-readable storage memory as recited in claim 15, wherein, in said transforming the plurality of separate microimages according to the transformation matrix, the program instructions are computer-executable to implement aligning one or more of the microimages according to one or more alignment parameters indicated by the transformation matrix and scaling one or more of the microimages according to one or more scale parameters indicated by the transformation matrix. 19. The computer-readable storage memory as recited in claim 18, wherein, in said transforming the plurality of separate microimages according to the transformation matrix, the program instructions are computer-executable to implement cropping one or more of the microimages according to one or more crop parameters indicated by the transformation matrix. 20. The computer-readable storage memory as recited in claim 15, wherein, in said applying a calibration technique to the radiance image to generate a transformation matrix for the radiance image, the program instructions are computer-executable to implement: receiving alignment input via the user interface specifying one or more alignment parameters for microimages in the radiance image;receiving scaling input via the user interface specifying one or more scale parameters for microimages in the radiance image;receiving cropping input via the user interface specifying one or more crop parameters for microimages in the radiance image; andgenerating the transformation matrix for the radiance image according to the alignment parameters, scale parameters, and crop parameters.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.