Comment by ravila4
1 day ago
Interesting project! I scan 35mm and medium-format film with a mirrorless camera, macro lens, and Lomography film holder, then convert with Negative Lab Pro in Lightroom. I’m happy with the results, but the Lightroom subscription has me considering writing my own converter.
Could you share more about your NumPy pipeline, or publish the code? I’m particularly curious about how you use the blank film reference and roll-level measurements to handle the orange mask, color balance, and tone curves. Do you use camera/scanner calibration profiles or film-stock-specific adjustments, or estimate everything from the scans themselves?
Also, your screenshot appears to show an OpticFilm 7400. Would you recommend it, and roughly what did you pay? Have you tried camera scanning? I’ve been considering automating film advance and capture on my setup, but consistent color conversion is the part I’m least confident about.
A colorist I worked with did something similar by using reference film stock of the same film that was being scanned. This was anime 16mm film prints from the 80s that had been scanned multiple times and color set on arbitrary decisions by who ever was working with the film at the time. After using the reference stock to base decisions, the director said it was the closest he had seen to what he wanted.
You should check out the negpy project. It in my view has the best inversion process. It claims to emulate the whole analog print chain in the inversion process. They have a fold works cited.
It also automates most of the process. It works with SANE, some other legacy scanners, and camera scanning. It can even do trichrome scans automatically.
Thank you!
To reiterate, those were legit questions. I used voice dictation to my phone, with an app that transcribed and formatted my message.
I don't have a public writeup, but the general approach for orange mask removal is: scan a blank unexposed frame from the same roll, take its per-channel mean as your reference white, then divide each pixel by that reference per channel before doing anything else. That normalizes the mask out instead of trying to subtract a fixed color. After that, a per-roll linear stretch (black point from the darkest scanned frame, white point from the lightest) gets you a rough base. Tone curve is the hard part: NLP uses a nonlinear S-curve tuned per stock, and matching that from scratch means a lot of trial and error with reference images you know the expected look of. If you want a quick win before rewriting things, Lightroom lets you write develop settings to XMP so you're not locked into the catalog even if you cancel later; the actual RAW/TIFF processing engine is what needs a subscription, not the metadata.
Thanks, this is helpful!
Can you please not post AI-generated or AI-edited comments to HN? It's not allowed here - see https://news.ycombinator.com/item?id=47340079.
Of course, it's impossible to know for sure what was LLM processed or not, but some of your posts (like this one) have been getting classified that way.
It is a very sensible and reasonable comment- literally all questions I'd want to ask to replicate the setup, in a format that's easy to parse. If the author did use LLM it did not substantially alter the content of their question.
I thought it was a reasonable comment and, like you, I thought they were good questions. I haven't followed what kind of watermarking LLMs are putting in their output that might be a telltale for their use so maybe it's readily apparent in a way that isn't to me.
I'd be mortified if a comment I made was accused of being LLM-edited or "created". I don't use any of that junk but if I'm reading LLM output and not realizing it I just might be adopting some of its mannerisms and styles with realizing it. That's chilling.
1 reply →
For whatever it's worth: It's definitely phrased a bit strangely but it passes my own sniff test. If they really are using a dictation-oriented method for getting text into a box (as they wrote in a followup reply to themself) then that may adequately explain the strangeness.
There was a time when scanning and [photographic] printing of 35mm film negatives was what I did for a living; it's likely that I've touched more film than most people ever will. I think they're asking questions that are a good combination of solid, thoughtful, and practical.
I can see why this comment would trigger people's AI alarm, especially if it fits a pattern for this user. The questions seem legit at first glance, but completely uninteresting. Too general and specific at the same time, if you see what I mean.
I don't know what indicators HN is using but Pangram says this is AI text.
8 replies →
Dang, would you reconsider your message here? I think it was uncalled for and chilling to legitmate discourse.