Perform Content Analysis Using Dolby Vision™ Content Mapping

Once the HDR track is created, you are ready to analyze the shots of the sequence using Dolby Vision technology.

This analysis, also known as Level 1 Analysis Metadata for Dolby Vision, finds the minimal, maximal, and average luminance values of the content needed to create the HDR/SDR result for the target display, which can be previewed using the eCMU or the iCMU.

When you are asked to deliver L1 metadata, it means that you must perform a content analysis of the shots in the sequence.

Important: If you work with letter boxed/pillar boxed content, make sure to define the Active Image Area to match your content framing before performing analysis or the L1 values will be wrong. More detail in Working with Letterbox and Pillarbox.

To create L1 metadata for a shot:

  1. Create the HDR track.

  2. In the Effects environment, display the HDR track.

    Change track with the Track and Version Selection box on the left, above the HDR Mode section.

  3. Select the HDR shot.

  4. Select a Target Display.

    The results of the analysis are biased by the Mastering display. You must always perform an analysis for the 100-nit Rec. 709 display. This is a requirement for any Dolby Vision XML deliverable.

  5. Set Analysis Scope to Current Shot.

  6. Click Analyze.

    Once the analysis completes, the HDR indicator on the shot turns white to indicate a completed analysis. The Analysis Result shows the Minimal, Maximal, and Average luminance values.

You can set the scope of the analysis with the Analysis Scope box:

If you encounter invalidation notifications:

To discard the results of the analysis:

  1. Select the shots and click Reset.

The analysis is very simple, but there are some limitations:

Working with Letterbox and Pillarbox

If you are working with shots containing contains letterbox / pillarbox, you must exclude those areas from the analysis. Use the Active Image Area to do just this: this ensures the analysis and the trims only include relevant pixels.