New post: Librem 5 Photo Processing Tutorial

During the last year, it has become possible to take some pretty good-looking high resolution photos with the Librem 5:

Both cameras can be already used to take stunning photos, and the driver for the front camera is even already upstreamed into mainline Linux. The processing pipeline is, however, rather limited at this point. The user has to manually set the gain, exposure and focus knobs before taking the photo, and there are no heuristics that could analyze the image and decide how to proceed with processing to make it look best.


Developing the raw picture automatically is currently done via dcraw tool, which in my experience can provide excellent results for some kind of photos, but can also come up with rather underwhelming pictures for other kinds. Nothing’s lost though – we can still take the raw DNG file and develop it manually, which is what I’ve done for most of my photos posted on the fediverse and what I’d like to demonstrate in this post.

Read more:


Regarding the little furry chair thief, I recommend RAIC technology (Redundant Array of Independent Chairs). That way, when chair theft occurs, you can wheel the chair away and wheel in a spare.


It’s very impressive what a small team like Librem has managed to achieve in many areas including the camera. I remember the first images a year ago, in black and white, terrible.

I understand that to get such images, there are still a lot of manual adjustments but still.

1 Like

I see a major weakness of this solution:



Reading that article makes me appreciate how incredible automatic image processing is on the average smartphone. I never realized just how complex cell phones were until I started following the development of the Librem 5 and PinePhone. The amount of collective labor to make a phone that is easy to use is truly staggering. I wish that Purism had the funding to pay for a hundred programmers rather than ten to work on the L5, because there is so much work to be done.


Hey @dos, you’re famous! (Or more famous.) :slight_smile:


I’d say the cat is more famous, as more look at pics than read, but each their own, and so on :slight_smile:


The article is awesome, because it shows at least 2 things:

  1. The hardware is capable enough to make good photos.
  2. If one has the skills and patience, a good photo is possible even at the moment.

On the other hand the whole experience is so inefficient. It is like chopping wood and staring the fireplace to get warm in the winter. Yes, it could be a fun way to spend time and it has its charm, but if the goal is simply to get warm ( or to take a photo) it is definitely awfully inefficient and it wastes tons of valuable human time compared to just switching on the central heating (taking photo on a device that would do most of the work on itself for 99% of the users).
I don’t want to spend an hour of my life in order to shoot one photo.
As I already wrote in another post, the camera is most important for video calls and for scanning QR codes as these are the communication and utility driven functions of a device intended to communicate with other people and communication is not only voice and text, but also video. And this article does not give much hope for getting video calls (in any apps) soon.

Sometimes I wonder if there could have been a different approach to Librem 5 - for example no camera and no WiFi, but several USB-s so that one could connect such external devices as dongles when needed.
YubiKey Nano C for example could have been a substitute for the Smart Card reader or one could use Librem Key per USB.
The micro SD is also not so comfortable when linked to the SIM card and I tend to use micro SD cards via a docking station for easy mount and unmount during use.
I know that this would have been very power consuming and that there might be other issues. I am just brainstorming aloud.

I think this is relatively undemanding as far as post-processing is concerned. You don’t need to spend 1 hour working on brightness, dynamic range and colour balance to get a good enough image of a QR code. The QR code even makes it extra easy by having built-in redundancy.

I understand that you are just brainstorming aloud but I really think this depends on your use case.

For me the uSD card just extends the available internal storage. It is almost never removed or inserted. So an internal card reader is what I want (with a high capacity uSD card in it).

For other people, yep, they might want a set of compact USB-C flash drives and plug in / swap over as needed. (For more heavy duty use, you might want a compact USB-C portable SSD, which I have tested with my Librem 5 and works fine.)

1 Like

Thank you for this tutorial. I wish I’d seen this before I started taking pics with the phone. Some of my pictures have been too bright, with washed out colors. Today, I resorted to putting my sunglasses in front of the lens, which actually helped. LOL. The next time I take a picture, I can use what I learned in the tutorial.

Very nice. Here is my pic taken in a brightly lit restaurant, with a light overhead, and ‘torch’ On. I already showed the comparison between my device taking a picture, and a old flip phone.