Conversation
libcamera@floss.social

libcamera

The SoftISP is making progress, allowing libcamera to operate on platforms where the ISP (Image Signal Processor) is either unusable, or not available.

Here we have Hans comparing the Intel IPU6 running the out-of-kernel stack against the libcamera SoftISP. The colours are off, as the white balance isn't handled yet - and there is no denoise.

But clear progress to functioning cameras!

4
5
1
mupuf@fosstodon.org

Martin Roukala (né Peres)

Reply to @libcamera@floss.social

@libcamera Congrats! And congrats to Hans for the majestic stare :D

Too bad I can't make it to FOSDEM this year, would have been nice to hang out once again, it's been a while!

0
0
0
NetworkManager@fosstodon.org

NetworkManager

Reply to @libcamera@floss.social

@libcamera needs a few more iterations for Hans' face to brighten up sufficiently
❤️

0
0
1
pavel

Pavel Machek

Reply to @libcamera@floss.social
@libcamera I have patches to get that working on PinePhone. And yes, it is step forwards, but there's still huge ammount of work ahead. This is AE (auto-exposure), and AE is quite easy. AF is harder (and missing from this). Plus, one day we'll want video recording. That's hard. And we'll want to take photos. That's also quite hard. Millipixels has some parts, but more should go in libcamera, and we really need drivers for jpeg/mpeg encoding.
1
0
4
rmader@floss.social

Robert Mader

Reply to @pavel

@pavel @libcamera accelerated encoding is somewhat independent of this, no? I.e. if the a device supports encoding via V4L2 or VA-API, that can be handled by applications / frameworks like Gstreamer - and doesn't need to live in , right?

1
0
1
pavel

Pavel Machek

Reply to @rmader@floss.social
@rmader @libcamera It does not need to live in libcamera, but what you ultimately want is good quality video, and there a lot of integration will be required. You want different AE/AF while recording video. Plus you need (for example) to tune debayer well to work nicely with the video encoding, yet leave enough resources for the encoding pipeline.
0
0
0
tomeu@mastodon.social

Tomeu Vizoso

Reply to @libcamera@floss.social

@libcamera Do you know if the denoising could be performed with a neural network?

1
0
0
libcamera@floss.social

libcamera

Reply to @tomeu@mastodon.social

@tomeu well that's something I'd be asking you about :-) looking forward to trying out your work sometime.

0
0
0
About social.kernel.org

Terms of service

  1. Please do not use this service in violation of the Linux Kernel Code of Conduct. Doing so will result in your account suspension with the referral of the matter to the CoC committee.

  2. "Repeating"/"boosting" someone else's status on this platform will be treated as endorsement and will fall under rule #1.

  3. You are encouraged to use this platform to promote your work on the Linux Kernel, but there is no restriction on permitted topics (with the exception of anything covered by #1 above).

  4. There is no requirement to post in English, but it should be considered the primary language of communication on this platform.

Privacy notice

The admins of this service have access to all posted statuses. They aren't looking, but if it's something they shouldn't know about, then you should not post it on this platform.

Please see the Linux Foundation Privacy Policy, which applies to this platform as well.

Getting your own account

If you would like an account on this instance, please check that the following applies to you:

  • You are listed in MAINTAINERS or CREDITS

  • OR: You have a kernel.org account or email address

  • OR: You have a long and established history of involvement with the Linux Kernel

If the above is true and you agree with the Terms of Service and Privacy Notice listed above, please use these instructions to request an account: