Change Details

Problem ====== SMPTE ST.2086 requires HDR metadata to be implemented, of which two elements seem to be unique to the image data being encoded: From [[ https://www.linkedin.com/pulse/hdr-10-metadata-smpte-st2086-maxfall-maxcll-carlos-carmona | here ]]: > Maximum Frame--Average Light Level (MaxFALL) is additional static HDR metadata that corresponds to the highest frame average brightness per frame in the entire stream. The MaxFALL metadata value is represented in nits, a unit of luminance. > > Maximum Content Light Level (MaxCLL) is additional static HDR metadata that corresponds to the brightest pixel in the entire stream. The MaxCLL metadata is represented in nits. > > MaxFALL/MaxCLL metadata is calculated on active picture only. Black mattes are not processed during the calculation. Solution: ---------- Right now we have a manual input for these values. The AVG fall data can be calculated from LODs, but the maxcll would be less precise(due to LODs being resized). The other option is to have an extra pass while rendering. @dkazakov wants to have this metadata stored in the KRA file, question is 'how should it look', as well as a bit of GUI.

Problem ====== SMPTE ST.2086 requires HDR metadata to be implemented, of which two elements seem to be unique to the image data being encoded: From [[ https://www.linkedin.com/pulse/hdr-10-metadata-smpte-st2086-maxfall-maxcll-carlos-carmona | here ]]: > Maximum Frame--Average Light Level (MaxFALL) is additional static HDR metadata that corresponds to the highest frame average brightness per frame in the entire stream. The MaxFALL metadata value is represented in nits, a unit of luminance. > > Maximum Content Light Level (MaxCLL) is additional static HDR metadata that corresponds to the brightest pixel in the entire stream. The MaxCLL metadata is represented in nits. > > MaxFALL/MaxCLL metadata is calculated on active picture only. Black mattes are not processed during the calculation. Solution: ---------- Right now we have a manual input for these values. The AVG fall data can be calculated from LODs, but the maxcll would be less precise(due to LODs being resized). The other option is to have an extra pass while rendering. @dkazakov wants to have this metadata stored in the KRA file(For headless rendering), question is 'how should it look', as well as a bit of GUI. This is relevant for both animations and still images, for example, the case of a matte painting or overpaint over a video still.

Problem ====== SMPTE ST.2086 requires HDR metadata to be implemented, of which two elements seem to be unique to the image data being encoded: From [[ https://www.linkedin.com/pulse/hdr-10-metadata-smpte-st2086-maxfall-maxcll-carlos-carmona | here ]]: > Maximum Frame--Average Light Level (MaxFALL) is additional static HDR metadata that corresponds to the highest frame average brightness per frame in the entire stream. The MaxFALL metadata value is represented in nits, a unit of luminance. > > Maximum Content Light Level (MaxCLL) is additional static HDR metadata that corresponds to the brightest pixel in the entire stream. The MaxCLL metadata is represented in nits. > > MaxFALL/MaxCLL metadata is calculated on active picture only. Black mattes are not processed during the calculation. Solution: ---------- Right now we have a manual input for these values. The AVG fall data can be calculated from LODs, but the maxcll would be less precise(due to LODs being resized). The other option is to have an extra pass while rendering. @dkazakov wants to have this metadata stored in the KRA file(For headless rendering), question is 'how should it look', as well as a bit of GUI. This is relevant for both animations and still images, for example, as well as a bit of GUIthe case of a matte painting or overpaint over a video still.