Systems and methods for non-linear processing of image frames

Abstract

Embodiments of the present disclosure include systems and methods for image processing for enhancing, e.g., photos, videos, and graphics representations. Particularly, some embodiments perform a histogram-based SMQT decomposition of various attributes in an original image. The SMQT decomposition of these attributes may then be used to create a look-up-table indicating how to modify pixel attributes in the same or a subsequent image so as to enhance the same or subsequent image. The results of the SMQT decomposition for each attribute may be interpolated based, e.g., on user preferences, to achieve a more optimal enhancement.

1 . A computer-implemented method comprising:
 obtaining a first image;   generating a first histogram for a first attribute associated with pixels in the first image;   generating a second histogram for a second attribute associated with pixels in the first image;   generating a first Successive Mean Quantization Transform (SMQT) output using the first histogram; and   generating a second SMQT output using the second histogram; and   generating an interpolation look-up-table based at least in part upon the first SMQT output, the second SMQT output, and an interpolation parameter; and   adjusting a pixel in a second image using the interpolation look-up-table.   
     
     
         2 . The computer-implemented method of  claim 1 , wherein generating a first SMQT output comprises generating an SMQT table for a first attribute. 
     
     
         3 . The computer-implemented method of  claim 1 , wherein the first attribute is one of a red, green, or blue pixel value attribute and the second attribute is a luminance value attribute. 
     
     
         4 . The computer-implemented method of  claim 1 , wherein the first attribute comprises a red pixel value attribute and the second attribute comprises a luminance pixel value attribute, the method further comprising:
 generating a third histogram for a green pixel value attribute;   generating a fourth histogram for a blue pixel value attribute;   generating a third SMQT output using the third histogram; and   generating a fourth SMQT output using the fourth histogram, and wherein   the generation of the interpolation look-up-table is further based upon the third SMQT output and the fourth SMQT output, the interpolation look-up-table specifying an RGB output value for a given RGB input value.   
     
     
         5 . The computer-implemented method of  claim 1 , wherein the first SMQT output is derived from an SMQT comprising 8 levels and the second SMQT output is derived from an SMQT comprising 8 levels. 
     
     
         6 . The computer-implemented method of  claim 1 , wherein generating a first histogram for a first attribute associated with pixels in the first image comprises iterating over less than all of the pixels in the image. 
     
     
         7 . The computer-implemented method of  claim 1 , wherein the interpolation parameter is received from a human user. 
     
     
         8 . A non-transitory computer-readable medium comprising instructions configured to cause at least one processor to perform a method, comprising:
 obtaining a first image;   generating a first histogram for a first attribute associated with pixels in the first image;   generating a second histogram for a second attribute associated with pixels in the first image;   generating a first SMQT output using the first histogram; and   generating a second SMQT output using the second histogram; and   generating an interpolation look-up-table based at least in part upon the first SMQT output, the second SMQT output, and an interpolation parameter; and   adjusting a pixel in a second image using the interpolation look-up-table.   
     
     
         9 . The non-transitory computer-readable medium of  claim 8 , wherein generating a first SMQT output comprises generating an SMQT table for the first attribute. 
     
     
         10 . The non-transitory computer-readable medium of  claim 8 , wherein the first attribute is one of a red, green, or blue pixel value attribute and the second attribute is a luminance value attribute. 
     
     
         11 . The non-transitory computer-readable medium of  claim 8 , wherein the first attribute comprises a red pixel value attribute and the second attribute comprises a luminance pixel value attribute, the method further comprising:
 generating a third histogram for a green pixel value attribute;   generating a fourth histogram for a blue pixel value attribute;   generating a third SMQT output using the third histogram; and   generating a fourth SMQT output using the fourth histogram, and wherein   the generation of the interpolation look-up-table is further based upon the third SMQT output and the fourth SMQT output, the interpolation look-up-table specifying an RGB output value for a given RGB input value.   
     
     
         12 . The non-transitory computer-readable medium of  claim 8 , wherein the first SMQT output is derived from an SMQT comprising 8 levels and the second SMQT output is derived from an SMQT comprising 8 levels. 
     
     
         13 . The non-transitory computer-readable medium of  claim 8 , wherein generating a first histogram for a first attribute associated with pixels in the first image comprises iterating over less than all of the pixels in the image. 
     
     
         14 . The non-transitory computer-readable medium of  claim 8 , wherein the interpolation parameter is received from a human user. 
     
     
         15 . A computer system, comprising:
 one or more hardware modules configured to:
 receive a first image; 
 generate a first histogram for a first attribute associated with pixels in the first image; 
 generate a second histogram for a second attribute associated with pixels in the first image; 
 generate a first SMQT output using the first histogram; and 
 generate a second SMQT output using the second histogram; and 
 generate an interpolation look-up-table based at least in part upon the first SMQT output, the second SMQT output, and an interpolation parameter; and 
 adjust a pixel in a second image using the interpolation look-up-table. 
   
     
     
         16 . The computer system of  claim 15 , wherein generating a first SMQT output comprises generating an SMQT table for the first attribute. 
     
     
         17 . The computer system of  claim 15 , wherein the first attribute is one of a red, green, or blue pixel value attribute and the second attribute is a luminance value attribute. 
     
     
         18 . The computer system of  claim 15 , wherein the first attribute comprises a red pixel value attribute and the second attribute comprises a luminance pixel value attribute, the method further comprising:
 generating a third histogram for a green pixel value attribute;   generating a fourth histogram for a blue pixel value attribute;   generating a third SMQT output using the third histogram; and   generating a fourth SMQT output using the fourth histogram, and wherein   the generation of the interpolation look-up-table is further based upon the third SMQT output and the fourth SMQT output, the interpolation look-up-table specifying an RGB output value for a given RGB input value.   
     
     
         19 . The computer system of  claim 15 , wherein the first SMQT output is derived from an SMQT comprising 8 levels and the second SMQT output is derived from an SMQT comprising 8 levels. 
     
     
         20 . The computer system of  claim 15 , wherein generating a first histogram for a first attribute associated with pixels in the first image comprises iterating over less than all of the pixels in the image. 
     
     
         21 . The computer system of  claim 15 , wherein the interpolation parameter is received from a human user.