High Density Encoding for ARRIRAW Image files

Codex High Density Encoding (HDE) is an encoding and decoding schema for large format ARRIRAW workflows to control the ever-increasing data footprint of RAW data. Studios are demanding 4K and RAW image data to archive and future-proof the negative. High Frame Rate (HFR) and High Dynamic Range (HDR) add to this large format data footprint, leading to significant cost increases as well as storage and transmission challenges. Uncompressed RAW data costs more to transport and store than compressed ProRes files. An uncompressed ARRIRAW image can be reduced with HDE to a file size not much larger than a corresponding ProRes 4444 XQ file. Except with Codex HDE, you can access the original pixel values that were encoded. Codex HDE can be used to compress raw Bayer pattern data (e.g. from a digital image sensor) in such a way that decoded data is identical to the raw data. The encoded files are typically 50-60% of their original size. A wide variety of third-party commercial applications can be used to decode files containing HDE bitstreams for data management, transcoding, color grading, visual effects and archiving. The presentation will describe the structure of data encoded with Codex HDE, along with algorithms for decoding the stored data. It is the intent of the presentation to describe the structure and encoding of all fields in the HDE bitstream, detailing how 3rd party vendors and their applications can decode HDE bitstreams and correctly identify its structure. Recent details on SMPTE RDD51 on HDE will be presented.

Published

2020-11

Content type

Original Research

Keywords

CODEX, ARRI, ARRIRAW, RAW, HDE, High Density Encoding, ZigZag Encoding, RDD 30:2014, RDD 51:2020

DOI

10.5594/M001908