Search

EP-4738813-A1 - LIMITATION OF A QUANTIZATION EFFECT WHEN EXCHANGING VIDEO CONTENT USING A CHANNEL WITH DIFFERENT COLOR GAMUTS BY YUV ADJUSTMENT

EP4738813A1EP 4738813 A1EP4738813 A1EP 4738813A1EP-4738813-A1

Abstract

A method comprising: obtaining (900) YUV data resulting from a conversion of a YUV signal with a first transfer function and a first color gamut to a YUV signal with a second transfer function and a second color gamut; and applying a correction to a component of the YUV data to limit effects of quantization errors.

Inventors

  • LE NAOUR, Robin
  • TOUZE, DAVID
  • MORVAN, PATRICK
  • CARAMELLI, NICOLAS

Assignees

  • InterDigital CE Patent Holdings, SAS

Dates

Publication Date
20260506
Application Date
20241031

Claims (14)

  1. A method comprising: obtaining (900) YUV data resulting from a conversion of a YUV signal with a first transfer function and a first color gamut to a YUV signal with a second transfer function and a second color gamut; converting (901) the YUV data in first RGB data with the first transfer function and the first color gamut; converting (902, 903) a quantized version of the YUV data into second RGB data with the first transfer function and the first color gamut; determining (904) an error vector between the first and the second RGB data; selecting (905) a channel of the first RGB data to correct from respective values of channels of the first RGB data; determining (906) a channel correction matrix from partial derivatives of a function allowing converting a YUV signal with the second transfer function and the second color gamut in a RGB signal with the first transfer function and the first color gamut; determining (907) a correction vector representing a correction per channel of the YUV data from the channel correction matrix and from a component of the error vector depending on the selected channel of the first RGB data to correct; determining (908) one channel of the YUV data to be corrected from a smallest value of component of the correction vector; and applying a correction corresponding to the smallest value to the determined channel of the YUV data.
  2. The method of claim 1 wherein, during the selecting (905) of a channel of the first RGB data to correct from respective values of channels of the first RGB data, the selected channel corresponds to the channel of the first RGB data having a lowest value.
  3. The method of claim 1 or 2 wherein the partial derivatives are computed with respect to channels Y, U and V of the YUV data.
  4. The method of any previous claim wherein the selecting of a channel of the first RGB data to correct from respective values of channels of the first RGB data considers a relative position of the first RGB data from boundaries of the first color gamut.
  5. The method of any previous claim wherein the determining (907) of the correction vector comprises dividing the component of the error vector corresponding to the selected channel of the first RGB data to correct by a column of the channel correction matrix corresponding to the selected channel of the first RGB data to correct.
  6. The method of any previous claim wherein the second color gamut is larger than the first color gamut.
  7. A device comprising electronic circuity configured for: obtaining (900) YUV data resulting from a conversion of a YUV signal with a first transfer function and a first color gamut to a YUV signal with a second transfer function and a second color gamut; converting (901) the YUV data in first RGB data with the first transfer function and the first color gamut; converting (902, 903) a quantized version of the YUV data into second RGB data with the first transfer function and the first color gamut; determining (904) an error vector between the first and the second RGB data; selecting (905) a channel of the first RGB data to correct from respective values of channels of the first RGB data; determining (906) a channel correction matrix from partial derivatives of a function allowing converting a YUV signal with the second transfer function and the second color gamut in a RGB signal with the first transfer function and the first color gamut; determining (907) a correction vector representing a correction per channel of the YUV data from the channel correction matrix and from a component of the error vector corresponding to the selected channel of the first RGB data to correct; determining (908) one channel of the YUV data to be corrected from a smallest value of component of the correction vector; and applying a correction corresponding to the smallest value to the determined channel of the YUV data.
  8. The device of claim 7 wherein, during the selecting (905) of a channel of the first RGB data to correct from respective values of channels of the first RGB data, the selected channel corresponds to the channel of the first RGB data having a lowest value.
  9. The device of claim 7 or 8 wherein the partial derivatives are computed with respect to channels Y, U and V of the YUV data.
  10. The device of any previous claim from claim 7 to 9 wherein the selecting of a channel of the first RGB data to correct from respective values of channels of the first RGB data considers a relative position of the first RGB data from boundaries of the first color gamut.
  11. The device of any previous claim from claim 7 to 10 wherein the determining (907) of the correction vector comprises dividing the component of the error vector corresponding to the selected channel of the first RGB data to correct by a column of the channel correction matrix corresponding to the selected channel of the first RGB data to correct.
  12. The device of any previous claim from claim 7 to 11 wherein the second color gamut is larger than the first color gamut.
  13. A computer program comprising program code instructions for implementing the method according to any previous claim from claim 1 to 6.
  14. Non-transitory information storage medium storing program code instructions for implementing the method according to any previous claims from claim 1 to 6.

Description

1. TECHNICAL FIELD At least one of the present embodiments generally relates to the field of production of video and more particularly to a method, a device and a system for limiting effects of quantization in a conversion of a video content from a first color gamut to a second color gamut. 2. BACKGROUND In a typical video system, many different video devices are interconnected to exchange video data. However, these devices may be designed to use different formats. A format conversion is therefore required to insure interoperability between the various devices. For example, the recent appearance of HDR (High Dynamic Range) systems offering video contents in a dynamic range greater than that of standard-dynamic-range video (SDR video) contents creates a need for such format conversion. Indeed, in the next years, HDR systems will coexists with SDR systems which implies a need for converting HDR video contents in SDR format and conversely SDR video contents in HDR format. A SDR video content is typically using "8" bits or "10" bits YUV data with a BT.709 Opto-electrical transfer function (OETF) and a BT.709 color gamut, as described in the BT.709 recommendation (Recommendation ITU-R BT.709-6, Parameter values for the HDTV standards for production and international program exchange, 06/2015) A HDR video content is typically using "10" bits or "12" bits YUV data with PQ or HLG Opto-electrical transfer function and BT.2020 color gamut as described in BT.2100 recommendation (Recommendation ITU-R BT.2100-2, Image parameter values for high dynamic range television for use in production and international program exchange, 07/2018). In the digital domain, exchanged video data are generally quantized data, the quantization being introduced at least by a binary representation of original data. A color gamut conversion scheme comprises several operations performed in the real domain, i.e. in the set of real numbers R (or at least with a precision larger than the precision of the quantized data (i.e. using a floating-point domain)). For instance, converting a "8" bits or "10" bits YUV video content with a BT.709 OETF and a BT.709 color gamut into a "10" bits or "12" bits YUV video content with a PQ or HLG OETF and a BT.2020 color gamut (and vice versa) comprises a conversion from a quantized domain (for example "8" or "10" bits domain) to a real domain and then a conversion from the real domain to another quantized domain (for example "10" or "12" bits domain). It is known that quantization introduces errors. Some operations performed during color gamut conversion may amplify these errors. These amplified errors may be significant, in particular, when a converted video content is converted back in its initial color gamut. It is desirable to overcome the above drawbacks. It is particularly desirable to propose a method limiting effects of quantization in a conversion of a video content from a first color gamut to a second color gamut. 3. BRIEF SUMMARY In a first aspect, one or more of the present embodiments provide a method comprising: obtaining YUV data resulting from a conversion of a YUV signal with a first transfer function and a first color gamut to a YUV signal with a second transfer function and a second color gamut;converting the YUV data in first RGB data with the first transfer function and the first color gamut;converting a quantized version of the YUV data into second RGB data with the first transfer function and the first color gamut;determining an error vector between the first and the second RGB data;selecting a channel of the first RGB data to correct from respective values of channels of the first RGB data;determining a channel correction matrix from partial derivatives of a function allowing converting a YUV signal with the second transfer function and the second color gamut in a RGB signal with the first transfer function and the first color gamut;determining a correction vector representing a correction per channel of the YUV data from the channel correction matrix and from a component of the error vector depending on the selected channel of the first RGB data to correct;determining one channel of the YUV data to be corrected from a smallest value of component of the correction vector; and applying a correction corresponding to the smallest value to the determined channel of the YUV data. In an embodiment, during the selecting of a channel of the first RGB data to correct from respective values of channels of the first RGB data, the selected channel corresponds to the channel of the first RGB data having a lowest value. In an embodiment, the partial derivatives are computed with respect to channels Y, U and V of the YUV data. In an embodiment, the selecting of a channel of the first RGB data to correct from respective values of channels of the first RGB data considers a relative position of the first RGB data from boundaries of the first color gamut. In an embodiment, the determining of the correction