ST 2073-3:2015 - SMPTE Standard - VC-5 Video Essence — Part 3: Image Formats

This standard specifies extensions to the VC-5 Elementary Bitstream defined in SMPTE ST 2073-1 for representing R′G′B′(A) images, Y′C′
1
C′
2
(A) images without color difference component sub-sampling, and Color Filter Array (CFA) images including Bayer image formats. This standard specifies the default color space for R′G′B′ and Y′C′
1
C′
2
color values. — This standard extends SMPTE ST 2073-1 by adding new parameters and chunk elements: — The ImageFormat parameter specifies the type of image: R′G′B′(A), Y′C′
1
C′
2
(A), CFA, or Bayer; — The PatternWidth, PatternHeight, and ComponentsPerSample parameters specify the structure of CFA images, including Bayer images; — The MaxBitsPerComponent parameter specifies an upper bound on the number of bits needed to represent all values in the component arrays present in the bitstream; — The VendorSpecificData chunk element allows private data to be embedded in a VC-5 bitstream; — The InverseTransform, InverseTransform16, and InversePermutation chunk elements specify how to transform and reorder decoded component arrays into the same color space and order as originally presented to the encoder; — The UniqueImageIdentifier chunk element specifies the image sequence and position within the sequence of an image represented by a VC-5 bitstream. — This standard extends SMPTE ST 2073-1 by defining three stages for transforming decoded component arrays into images: inverse component transform, inverse component permutation, and image repacking process. — This standard extends SMPTE ST 2073-1 to specify levels of performance for decoder implementations.