Sveriges mest populära poddar

Color Mentor

Interview with Menno Dreischor

120 min • 16 april 2025

In this episode I interview Menno Dreischor, a chemical engineer based in Amsterdam. 

His company TIPb offers engineering solutions for a wide range of industries. Menno’s interest in film led him to develop an algorithm to help with restoring the faded color on old film scans. It analyses the pixel distribution of a supplied source and target image and generates a matching LUT. It is currently in a beta testing phase and has many potential applications from restoration, shot matching and creative look design. He also has a second algorithm that tries to predict the aspects of an image that will induce visual cognition effects such as simultaneous contrast, assimilation and others effects. In the interview we discuss both projects and their potential applications.

Menno’s Color matching examples:

https://www.youtube.com/@colormatchinghub

We discuss:

  • (00:00) - Intro.
  • (00:53) - Menno introduction.
  • (02:51) - Interview starts.
  • (04:08) - Background in chemistry.
  • (05:29) - Connection to film.
  • (06:09) - Unofficial Star Wars restoration.
  • (08:46) - Matlab.
  • (10:25) - Drug analyzer.
  • (13:49) - Color Matching Project.
  • (15:30) - The matching process.
  • (16:45) - Histogram matching.
  • (21:12) - Theoretical basis.
  • (27:38) - Link to look design parameters.
  • (31:15) - Does it use machine learning?
  • (32:52) - Parameterization.
  • (40:54) - Entropy.
  • (48:26) - Open-source vs closed source.
  • (56:29) - Integration into current workflows.
  • (01:07:01) - Will this tool help or hinder creativity?
  • (01:13:34) - Over emphasis on matching.
  • (01:15:14) - Need for test footage.
  • (01:19:51) - Matching examples.
  • (01:31:39) - Visual cognition introduction.
  • (01:32:25) - The significance of cognition.
  • (01:33:47) - Does it affect the matching project?
  • (01:37:10) - How was it implemented?
  • (01:38:50) - Kernel based approach.
  • (01:40:50) - Assimilation's relation to resolution.
  • (01:42:21) - Visual cognition as a summary.
  • (01:47:14) - Matching is always done in context.
  • (01:48:46) - Uses of the cognition algorithm.
  • (01:53:41) - Discussion about compression and noise.
  • (01:55:41) - Screwball ideas on streaming.
  • (01:57:29) - Wrap up.
  • (01:59:58) - Outro.
  • (02:00:25) - End.

Check out   www.patreon.com/ColorMentor/    for colorist training and mentoring. ★ Support this podcast on Patreon ★
00:00 -00:00