Difference between revisions of "AXIOM Beta/Raw Recording"
| Line 29: | Line 29: | ||
| [[File:Axiom_raw_hdmi_b-frame.jpg.jpg | 400px]] | | [[File:Axiom_raw_hdmi_b-frame.jpg.jpg | 400px]] | ||
|} | |} | ||
==Post Processing Principle== | |||
'''Version 1 Prototype:''' | |||
[[File:Axiom_beta_raw_recording_principle.png]] | |||
Revision as of 12:36, 6 January 2022
1 Overview
Simplified: The AXIOM Recorder is a small computer connected to the AXIOM Beta via HDMI, we use the HDMI video stream to transport our own custom raw image data from the camera. This stream needs to be repacked or post processed on the AXIOM Recorder to acquire the actual raw image sequence (RAW12 sequence). First software prototypes did this after recording, newer versions do it on-the-fly while recording.
Technical: On the AXIOM Beta we utilize a Full HD (1920 x 1080) HDMI output stream with twice the frame-rate to store 4 pixels (12 bit each) from the image sensor (sensels) in 2 RGB pixels (24 bit each) in the HDMI stream. After recording this data through a HDMI capture device like the Magewell HDMI USB3 Gen 2 on a PC, Laptop or Single Board Computer (SBC) the data needs to be post processed to obtain the original raw data from the image sensor again. The two HDMI images from a single raw image are called A-Frame and B-Frame.
2 Recording Principle
3 Pixel Packing V3 Structure
Traditional HDMI signal:
AXIOM Beta raw mode output:
Essentially from a 2x2 sensel block on the image sensor R, G1 go into the A frame and G2, B into the B frame of a HDMI stream.
Images from the HDMI stream look like this then:
|
|
4 Post Processing Principle
Version 1 Prototype:
