Method and apparatus for estimating depth of focused plenoptic data
Method and apparatus for estimating the depth of focused plenoptic data are suggested. The method includes: estimating the inherent shift of in-focus pixels of the focused plenoptic data; calculating a level of homogeneity of the pixels of the focused plenoptic data; determining the pixels of the focused plenoptic data which either have disparities equal to the inherent shift or belong to homogeneous areas, as a function of the level of homogeneity of the pixels of the focused plenoptic data; and estimating the depth of the focused plenoptic data by a disparity estimation without considering the determined pixels. According to the disclosure, the pixels of the focused plenoptic data which either have a disparity equal to the inherent shift or belong to a homogeneous area will not be considered for the estimation of the depth, which can reduce computational costs and at the same time increase accuracy of estimations for in-focus parts of the scene.
1. A method for estimating disparity of raw focused plenoptic data, comprising:
estimating an inherent shift of in-focus pixels of the raw focused plenoptic data, the inherent shift including a constant shift between samples of a same 3D point on different microlens images of the raw focused plenoptic data;
determining a level of homogeneity of pixels of the raw focused plenoptic data;
estimating a depth of the raw focused plenoptic data by a disparity estimation without considering the pixels of the raw focused plenoptic data which either have disparities equal to the inherent shift or which belong to homogeneous areas as a function of the level of homogeneity of the pixels of the raw focused plenoptic data.
2. The method according to claim 1 , wherein the inherent shift is estimated as a function of a distance between a microlens grid of a camera capturing the raw focused plenoptic data and an in-focus plane of a main lens of the camera, an aperture and a focal length of microlens of the camera.
3. The method according to claim 1 , wherein the estimation of the inherent shift comprises estimating a shift between samples of a same point on sub-aperture views of the raw focused plenoptic data.
4. The method according to claim 3 , the inherent shift is estimated as a function of an aperture of microlens of a camera capturing the raw focused plenoptic data.
5. The method according to claim 1 , wherein the level of the homogeneity of each pixel is determined by estimating a measure of homogeneity among all of its corresponding pixels, and assigning the estimated metric to all of the corresponding pixels.
6. The method according to claim 1 , wherein the level of the homogeneity of each pixel is determined by calculating standard deviations among all of its corresponding pixels on three color channels, and assigning the estimated metric to all of the corresponding pixels.
7. The method according to claim 1 , wherein the level of the homogeneity of each pixel is determined from a matrix of views of the raw focused plenoptic data to represent the light field of the views, by calculating a measure of homogeneity of the corresponding pixels of each pixel on the plurality of views in the matrix of views on three color channels.
8. The method according to claim 7 , wherein the matrix of view is determined by:
estimating from the raw focused plenoptic data a position of a center of each micro-lens of a plenoptic camera capturing unfocused plenoptic data; and
demultiplexing for all angular coordinates the corresponding view of the unfocused plenoptic data by extracting from every microlens image the pixel at the spatial coordinate with respect to the center of every micro-lens image, and putting the extracted pixel in the view considering the inherent shift.
9. An apparatus for estimating disparity of raw focused plenoptic data, comprising:
a first estimating unit configured to estimate the inherent shift of the raw focused plenoptic data;
a first determining unit configured to determine a level of homogeneity of pixels of the raw focused plenoptic data;
a second determining unit configured to determine the pixels of the raw focused plenoptic data which either have disparities equal to the inherent shift or belong to homogeneous areas, as a function of the level of homogeneity of the pixels of the raw focused plenoptic data; and
a second estimating unit configured to estimate a depth of the raw focused plenoptic data by a disparity estimation without considering the determined pixels.
10. The apparatus according to claim 9 , wherein the first estimating unit is configured to estimate the inherent shift by estimating a shift between samples of a same point on different microlens images of the raw focused plenoptic data.
11. The apparatus according to claim 10 , wherein the first estimating unit is configured to estimate the inherent shift as a function of the distance between the microlens grid of a camera capturing the raw focused plenoptic data and an in-focus plane of a main lens of the camera, an aperture and a focal length of microlens of the camera.
12. The apparatus according to claim 10 , wherein the first estimating unit is configured to estimate the inherent shift by estimating a shift between samples of a same point on sub-aperture views of the raw focused plenoptic data.
13. The apparatus according to claim 10 , wherein the second determining unit is configured to determine the pixels of the raw focused plenoptic data which either have disparities equal to the inherent shift or belong to homogeneous areas by a thresholding the estimation homogeneity metric values which comprises:
setting a threshold on the standard deviation for each color channel of an in-focus pixel;
comparing the estimated homogeneity metric value per color channel with the threshold for that color channel; and outputting a value of 0 if the estimated homogeneity metric of the each color channel is less than the threshold, and
outputting a value of 1 otherwise.
14. Computer program product which is stored on a non-transitory computer readable medium and comprises program code instructions executable by a processor for implementing the method according to claim 1 .