Live data from Hacker News

What an unprocessed photo looks like

maurycyz.com

291–300 of 418 posts

Re: What an unprocessed photo looks like

#291
For anyone that enjoyed this, pick up a book on digital image processing. The first chapters of most such books cover this, in almost step-by-step fashion. And then the books will usually start to venture into more classical machine learning stuff.

Re: What an unprocessed photo looks like

#292

For anyone that enjoyed this, pick up a book on digital image processing. The first chapters of most such books cover this, in almost step-by-step fashion. And then the books will usually start to venture into more classical machine learning stuff.

Do you have any specific book recommendations for a layman?

Re: What an unprocessed photo looks like

#293
post #280

Earlier quoted context omitted.

Not a complaint, but both the final images have poor contrast, lighting, saturation and colour balance, making them a disappointing target for an explanation of how these elements are produced from raw sensor data. But anyway, I enjoyed the article.

That’s because it requires much more sophisticated processing to produce pleasing results. The article is showing you the absolute basic steps in the processing pipeline and also that you don’t really want an image that is ‘unprocessed’ to that extent (because it looks gross).

No, the last image is the "camera" version of it- though it's not clear if he means the realtime processing before snapping the picture or with the postprocessing that happens right after. Anyway, we have no way to understand how far the basic-processed raw picture is from a pleasing or normal-looking result because a) the lighting is so bad and artificial that we have no idea of how "normal" should look; b) the subject is unpleasant and the quality "gross" in any case.

Re: What an unprocessed photo looks like

#295
post #249

Earlier quoted context omitted.

> Something that surprised me is that very little of the computation photography magic that has been developed for mobile phones has been applied to larger DSLRs. Perhaps it's because it's not as desperately needed, or because prior to the current AI madness nobody had sufficient GPU power lying around for such a purpose. Sony Alpha 6000 had face detection in 2014.

Sure, and my camera can do bird eye detection and whatnot too, but that's a very lightweight model running in-body. Probably just a fine-tuned variant of something like YOLO. I've seen only a couple of papers from Google talking about stacking multiple frames from a DSLR, but that was only research for improving mobile phone cameras. Ironically, some mobile phones now have more megapixels than my flagship full-frame…

A DSLR and mobile phone camera optimize for different things and can't really be compared.

Mobile phone camera's are severely handicapped by the optics & sensor size. Therefore to create a acceptable picture (to share on social media) they need to do a lot of processing.

DSLR and professional camera's feature much greater hardware. Here the optics and sensor size/type are important it optimize the actual light being captured. Additionally in a professional setting the image is usually captured in a raw format and adjusted/balanced afterwards to allow for certain artistic styles.

Ultimately the quality of a picture is not bound to it's resolution size but to the amount and quality of light captured.

Re: What an unprocessed photo looks like

#296
This post reminded me of the blog posts [0] regarding the "megapixels" camera app for the pinephone , written by Martijn Braam. For those interested it dives quite deep into the color profiling and noise reduction and more to make the pinephone camera usefull.

[0] https://blog.brixit.nl/tag/megapixels/

Re: What an unprocessed photo looks like

#297
> For comparison, here’s the image my camera produced from the same data:

Is it just me or does his version look a lot better? The camera version has brighter colors in some places (the star and lights), but it seems to be at the cost of detail and contrast in the subtler areas.

Re: What an unprocessed photo looks like

#298

I love posts that peel back the abstraction layer of "images." It really highlights that modern photography is just signal processing with better marketing. A fun tangent on the "green cast" mentioned in the post: the reason the Bayer pattern is RGGB (50% green) isn't just about color balance, but spatial resolution. The human eye is most sensitive to green light, so that channel effectively carries the majority of t…

The bit about the green over-representation in camera color filters is partially correct. Human color sensitivity varies a lot from individual to individual (and not just amongst individuals with color blindness), but general statistics indicate we are most sensitive to red light. The main reason is that green does indeed overwhelmingly contribute to perceptual luminance (over 70% in sRGB once gamma corrected: https:…

> Small neural networks I believe are the current state of the art (e.g. train to reverse a 16x16 color filter pattern for the given camera). What is currently in use by modern digital cameras is all trade secret stuff.

Considering you usually shoot RAW, and debayer and process in post, the camera hasn't done any of that.

It's only smartphones that might be doing internal AI Debayering, but they're already hallucinating most of the image anyway.

Re: What an unprocessed photo looks like

#300
Another tool to add to my arsenal of responses to people who claim either "no filter used" or "SOOC photo". Both of those may be true for some values of "no filter" or "straight out of camera" but they're not remotely describing the reality that any digital image is heavily manipulated before it leaves the camera. And that's ok! Our eyes are filters. Our brain is a filter. Photographic film and processing techniques are filters. The use of "no filter" and "SOOC" to imply capturing something unedited and therefore authentic is the artificial thing.
Post reply on HN