High dynamic range codecs

Patent metadata

Field	Value
Publication number	US-9210439-B2
Application number	US-201514617445-A
Country	US
Kind code	B2
Filing date	Feb 9, 2015
Priority date	Jan 23, 2006
Publication date	Dec 8, 2015
Grant date	Dec 8, 2015

Abstract

A method for encoding high dynamic range (HDR) images involves providing a lower dynamic range (LDR) image, generating a prediction function for estimating the values for pixels in the HDR image based on the values of corresponding pixels in the LDR image, and obtaining a residual frame based on differences between the pixel values of the HDR image and estimated pixel values. The LDR image, prediction function and residual frame can all be encoded in data from which either the LDR image of HDR image can be recreated.

First claim

The invention claimed is:

1. A method for encoding an input high dynamic range (HDR) image represented in a first color space, the method comprising:
accessing an input lower dynamic range (LDR) image represented in a second color space, the input LDR image corresponding to the input high dynamic range image, wherein the second color space is different than the first color space;
generating a second HDR image; said generating the second HDR image comprising color transforming the input HDR image from the first color space to a third color space;
generating a second LDR image; said generating the second LDR image comprising color transforming the input LDR image from the second color space to a fourth color space;
generating a prediction function based on the second HDR image and the second LDR image;
obtaining a predicted HDR image, said obtaining comprising applying the prediction function to the second LDR range image;
generating a residual image representing differences between pixel values in the predicted HDR image and corresponding pixel values in the second HDR image; and
encoding the second LDR image, the prediction function, and the residual image.

2. The method according to claim 1 further comprising:
generating a quantized residual image, said generating the quantized residual image comprising quantizing the residual image with a quantization factor; and
encoding the second LDR image, the prediction function, the quantized residual image, and the quantization factor.

3. The method of claim 1 , wherein the generating the prediction function comprises binning pixels in the second high dynamic range image.

4. The method of claim 3 , wherein the generating the prediction function further comprises analyzing the binned pixels by a technique comprising at least one of the following group: arithmetic average, median, average maximum and average minimum set as limiting values for the binned pixels, using a function that is nonlinear in the logarithmic domain, computing a centroid of a subset of the binned pixels.

5. The method of claim 1 , further comprising generating a new prediction function for at least two frames in a video sequence, wherein the video sequence comprises the input LDR image and the input HDR image.

6. The method of claim 1 , wherein encoding the prediction function comprises encoding a look up table.

7. The method of claim 1 , wherein the third color space and the fourth color space are perceptually uniform color spaces.

8. The method of claim 1 , wherein the second color space comprises an sRGB color space and the fourth color space comprises a lu′v′ color space.

9. The method of claim 1 , wherein the first color space comprises a XYZ color space.

10. The method of claim 1 , further comprising generating the input LDR image by applying one of a tone-mapping function or a gamut mapping function to the input HDR image.

11. A method for decoding a datastream the method comprising:
receiving an encoded image in a first dynamic range and a first color space, an encoded residual image in a second color space, and a prediction function;
generating a first image in the first dynamic range and the first color space, said generating the first image comprises decoding the encoded image;
generating a decoded residual image in the second color space, said generating the decoded residual comprises decoding the encoded residual image;
applying the prediction function to the first image to obtain a second image in a second dynamic range and the second color space; and
generating a third image, said generating the third image comprises combining the decoded residual image with the second image;
wherein the prediction function is based on the first image in the first dynamic range and a reference image with a higher dynamic range than the first dynamic range to which the first image corresponds.

12. The method of claim 11 , wherein applying the prediction function comprises mapping pixel values of the first image to pixel values of the second image using a look-up table.

13. The method of claim 11 , further comprising:
receiving a quantization factor; and
dequantizing the decoded residual image with the quantization factor prior to the generating the third image.

14. The method of claim 11 wherein the second color space comprises a perceptually uniform color space.

Assignees

• Max Planck Gesellschaft
• Max Planck Gesellschaft

Inventors

• Efremov Alexander
• Mantiuk Rafal
• Krawczyk Grzegorz
• Myszkowski Karol
• Seidel Hans-Peter

Classifications

• H04N19/33

  in the spatial domain · CPC title

• H04N19/186

  the unit being a colour or a chrominance component · CPC title

• H04N19/136

  Incoming video signal characteristics or properties · CPC title

• H04N19/50 Primary

  using predictive coding (H04N19/61 takes precedence) · CPC title

• H04N19/187

  the unit being a scalable video layer · CPC title