Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

HDR mezzanine #58

Open
jpiesing opened this issue Sep 27, 2023 · 1 comment
Open

HDR mezzanine #58

jpiesing opened this issue Sep 27, 2023 · 1 comment
Assignees

Comments

@jpiesing
Copy link

jpiesing commented Sep 27, 2023

There is an interest in creating test content for static and dynamic HDR. That in turn needs mezzanine content that is in HDR.

@haudiobe provided the following links on possible content.

In 3GPP, we collected a database of content including HDR and SDR.

Details on the sequences can be found in TR 26.955 here: https://portal.3gpp.org/desktopmodules/Specifications/SpecificationDetails.aspx?specificationId=3741

The sequences can be accessed here: https://dash-large-files.akamaized.net/WAVE/3GPP/5GVideo/ReferenceSequences/. All sequences include copyright statements and metadata.

An overview of the sequences are provided here: https://dash-large-files.akamaized.net/WAVE/3GPP/5GVideo/ReferenceSequences/sequences.csv

@rbouqueau mentioned the following;

The HDR sequences from Netflix I talked about:
https://opencontent.netflix.com/

It was also mentioned that Ateme and Harmonic may be able to generate (dynamic) HDR from SDR.

@nicholas-fr
Copy link
Collaborator

Initial HDR10 mezzanine streams have been generated and can be found here:
https://dash-large-files.akamaized.net/WAVE/Mezzanine/development/2024-01-02/

Files (each with corresponding JSON):
sol_L1_1920x1080@25_60.mp4
sol_L1_1920x1080@30_60.mp4
sol_M1_2560x1440@25_60.mp4
sol_M1_2560x1440@30_60.mp4
sol_O1_3840x2160@25_60.mp4
sol_O1_3840x2160@30_60.mp4
sol_O1_3840x2160@50_30.mp4
sol_O1_3840x2160@50_60.mp4
sol_O1_3840x2160@60_30.mp4
sol_O1_3840x2160@60_60.mp4

Based on updated HEVC test vector sparse matrix here (now including cud1, clg1, chd1 in addition to chh1):
https://docs.google.com/spreadsheets/d/1Bmgv6-cfbWfgwn7l-z0McUUI1rMjaWEwrN_Q30jaWk4/edit?usp=sharing

They are based on the Netflix Sol Levante asset found here (CC BY 4.0) - thanks to Romain for mentioning them:
http://download.opencontent.netflix.com.s3.amazonaws.com/SolLevante/hdr10/SolLevante_HDR10_r2020_ST2084_UHD_24fps_1000nit.mov
http://download.opencontent.netflix.com.s3.amazonaws.com/SolLevante/imf/SolLevante_IMF_DolbyVision_PQP3D65_UHD_24fps/CPL_0a207596-a0f5-458a-811e-a97a4286d00c.xml

Next steps:

  • Create HLG10 mezzanine streams using HLG source of EBU Croatia sequence.
  • Resolve issues found with HDR10 streams if any.
  • Create remaining missing HDR10 streams (including those for splicing tests: A1 / AD-A1 labelled streams).
  • Update mezzanine scripts for HDR metadata handling (ST2086, MaxFALL, MaxCLL) - initial streams have been generated using some hardcoded values.
  • Add dynamic metadata to HDR10 mezzanine streams (e.g. ST2094-10, ST2094-40).

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants