Method and apparatus for compressing or decompressing light field images
A method for compressing a set of images comprising a plenoptic image and at least one refocused image obtained from the plenoptic image for a given focus, by generating a set of data comprising a compressed version of the plenoptic image, some focusing parameter to build a refocused image from this plenoptic image and a compressed residual data to restore the refocused image in its plain quality. Accordingly, the set of images can be compressed in an efficient way, preserving the quality of the shoot using the focus chosen by the photographer.
1. A method comprising:
obtaining a compressed plenoptic image generated by compressing an original plenoptic image;
obtaining a refocusing parameter;
generating a first refocused image from the original plenoptic image using the refocusing parameter and generating a second refocused image from a decompressed plenoptic image using the refocusing parameter, the decompressed plenoptic image being obtained from the compressed plenoptic image;
obtaining residual data corresponding to a difference between the first refocused image and the second refocused image; and
outputting data comprising the compressed plenoptic image, the residual data and the refocusing parameter.
2. The method according to claim 1 , wherein the outputted residual data is compressed.
3. The method according to claim 2 , wherein a block DCT transformation algorithm is used to compress the residual data.
4. The method according to claim 2 , wherein a quantization algorithm is used to compress the residual data.
5. The method according to claim 2 , wherein an entropy encoding algorithm is used to compress the residual data.
6. The method according to claim 1 , further comprising:
storing in a memory outputted data.
7. The method according to claim 1 , wherein the original plenoptic image is compressed using vector quantization.
8. The method according to claim 1 , wherein the original plenoptic image is compressed using an algorithm that is based on using an anchor.
9. The method according to claim 1 , further comprising:
computing a depth map and/or disparity map based on the original plenoptic image; and
computing the refocusing parameter from the depth map and/or the disparity map and from one or more refocusing user parameters.
10. A non-transitory computer-readable storage medium on which is stored code of an executable program that, when executed, causes a computer to execute each of the following steps:
obtaining a compressed plenoptic image generated by compressing an original plenoptic image;
obtaining a refocusing parameter;
generating a first refocused image from the original plenoptic image using the refocusing parameter and generating a second refocused image from a decompressed plenoptic image using the refocusing parameter, the decompressed plenoptic image being obtained from the compressed plenoptic image;
obtaining residual data corresponding to a difference between the first refocused image and the second refocused image; and
outputting data comprising the compressed plenoptic image, the residual data and the refocusing parameter.
11. A method comprising:
obtaining a compressed plenoptic image generated by compressing an original plenoptic image;
obtaining a refocusing parameter;
obtaining residual data corresponding to a difference between a first refocused image and a second refocused image, wherein the first refocused image is generated from the original plenoptic image using the refocusing parameter and the second refocused image is generated from a decompressed plenoptic image using the refocusing parameter, the decompressed plenoptic image being generated by decompressing the compressed plenoptic image;
generating a refocused image from a decompressed plenoptic image using the refocusing parameter, the decompressed plenoptic image being obtained from the compressed plenoptic image; and
generating an enhanced image from the generated refocused image using the residual data.
12. A computing device comprising:
an obtaining module for obtaining a compressed plenoptic image generated by compressing an original plenoptic image;
an obtaining module for obtaining a refocusing parameter;
a generating module for generating a first refocused image from the original plenoptic image using the refocusing parameter and generating a second refocused image from a decompressed plenoptic image using the refocusing parameter, the decompressed plenoptic image being obtained from the compressed plenoptic image;
an obtaining module for obtaining residual data corresponding to a difference between the first refocused image and the second refocused image; and
an outputting module for outputting data comprising the compressed plenoptic image, the residual data and the refocusing parameter.
13. The device according to claim 12 , further comprising a compressing module for compressing the outputted residual data.
14. The device according to claim 13 , wherein a block DCT transformation algorithm is used to compress the residual data.
15. The device according to claim 13 , wherein a quantization algorithm is used to compress the residual data.
16. The device according to claim 13 , wherein an entropy encoding algorithm is used to compress the residual data.
17. The device according to claim 12 , further comprising a storing module for storing in a memory the outputted data.
18. The device according to claim 12 , wherein a compressing module for compressing the original plenoptic image uses vector quantization.
19. The device according to claim 12 , wherein a compressing module for compressing the original plenoptic image uses an algorithm that is based on using an anchor.
20. The device according to claim 12 , further comprising:
a computing module for generating a depth map and/or disparity map based on the original plenoptic image; and
a computing module for generating the refocusing parameter from the depth map and/or the disparity map and from one or more refocusing user parameters.
21. A computing device comprising:
an obtaining module for obtaining a compressed plenoptic image generated by compressing an original plenoptic image;
an obtaining module for obtaining a refocusing parameter;
an obtaining module for obtaining residual data corresponding to a difference between a first refocused image and a second refocused image, wherein the first refocused image is generated from the original plenoptic image using the refocusing parameter and the second refocused image is generated from a decompressed plenoptic image using the refocusing parameter, the decompressed plenoptic image being generated by decompressing the compressed plenoptic image;
a generating module for generating a refocused image from a decompressed plenoptic image using the refocusing parameter, the decompressed plenoptic image being obtained from the compressed plenoptic image, and for generating an enhanced image from the generated refocused image using the residual data.