Raspberry Pi has released a new camera, which can be used with your hardware. The camera uses a sony IMX477-AACK sensor, which has the following characteristics:
It uses IMX477, a BSI & stacked 1/2.3" 12MP sensor.
It's capable of full 12MP readout at 60fps with 10bit ADC, and 40fps with 12bit ADC.
https://www.sony-semicon.co.jp/products/common/pdf/IMX477-AACK_Flyer.pdf
https://www.raspberrypi.org/blog/new-product-raspberry-pi-high-quality-camera-on-sale-now-at-50/
Yeah I saw that this morning, this is super interesting
LOLOLOLOLOLOLOLOLOLOL I have to buy this thing asap!
(https://www.raspberrypi.org/app/uploads/2020/04/stupid3-500x375.jpg)
Proportion reminds me of this gem:
(https://img.kentfaith.de/cache/catalog/lense/2248-canon-tv-5200mm-f-14-800x800.jpg)
And yes, the cam is mounted!
Looks promising, and I see a lot of potential. Not for replacing DSLR cameras, but for things like machine vision, security cameras or... just a half-decent webcam.
Quote from: yourboylloyd on May 01, 2020, 04:31:12 PM
LOLOLOLOLOLOLOLOLOLOL I have to buy this thing asap!
Obviously, that picture was made for marketing reasons. This sensor is small: 6.29x4.71 mm active area. Compare to Canon's 22.5x15 mm (APS-C) or 36x24 mm (full frame). A 3x crop on your favourite APS-C camera (EOS M, 700D, 600D etc) is going to have roughly the same size (on the sensor: 7.5x5 mm). Here's a calculator: https://www.scantips.com/lights/fieldofview.html
FYI, the C mount accepts 1" sensors, or even larger:
Quote from: https://en.wikipedia.org/wiki/C_mount
C-mount lenses are built for the 8 mm and 16 mm film formats and the 1/3", 1/2", 2/3", 1", and 4/3" video formats, which corresponds to a range of image circles approximately from 5 to 22 mm in diameter.
But yes, you will be able to swap lenses between this thing and your Canon DSLR (with adapters). In both directions if you use an EOS M.
Anyway, don't underestimate the power of RPi user community. Pretty sure they have a quite a few highly skilled - active - developers ;)
https://www.techrepublic.com/article/raspberry-pi-why-sales-have-rocketed-in-the-middle-of-the-coronavirus-outbreak/
Edit - found some more details:
- Camera-Guide.pdf (https://magazines-static.raspberrypi.org/books/full_pdfs/000/000/036/original/Camera-Guide.pdf)
- 2028x1520 at 0.1-50 fps (2x2 binning)
- 4056x3040 at 0.005-10 fps (even if the sensor is faster, probably the bottleneck is on the RPi side)
- long exposures up to 200s (https://stargazerslounge.com/topic/353320-new-higher-res-camera-for-the-raspberry-pi/)
Wanted: complete sensor documentation, not just the 2-page datasheet :)
Quote from: a1ex on May 01, 2020, 05:59:44 PM
Not for replacing DSLR cameras, but for things like machine vision
I'm contributing to a dataset for a computer vision project that would highly benefit from it. I wonder if it's able to output Raw or if they have the firmware open sourced to tweak compression parameters. One issue we are having with the dataset images is that most cameras have too much compression and this affects quality of object labeling. Outputing raw would not just improve overall labelling but also open to possibilities of pre-processing those images (to decrease haze on low visibility scenarios, for example).
Previous camera modules are able to output raw image/video (didn't try):
https://github.com/6by9/raspiraw
https://www.raspberrypi.org/forums/viewtopic.php?f=43&t=109137
https://hackaday.com/2019/08/10/660-fps-raspberry-pi-video-captures-the-moment-in-extreme-slo-mo/
BTW - will this work on RPi Zero? Apparently it needs a different cable (https://coolcomponents.co.uk/products/raspberry-pi-zero-camera-cable).
any dng samples ? Id buy it if it
shoots raw video
https://www.google.com/amp/s/amp-blog.robertelder.org/recording-660-fps-on-raspberry-pi-cameravut all they have is crap resolutions,probably cause card reader is a bootleneck, they should start with that, and provide highest res info sheet in typical fps like 30 25 24 60 , not 660 with 64 pixels height.
I guess resolutions in 24 or 30 fps cant be too high , suspiciously they dont talk about it which could mean theyre poor not even hd.
I hope im wrong tho.
---
ok its 30 fps max in hd nobinning but crop and higher res or nocrop is 15 fps max, binned 2x nocrop mode is max 1640*1232 4:3 or 15:9 aspect ratio and 40 fps info in pdf is vague , unless someone proves otherwise .
Ill pass until its going to be higher than 1080 in 24fps.
Or crop from fullres to lower one to achieve 24ps maybe? Maybe worth a try
There are other options toshiba TC358840 Camera Serial Interface Converter Chipset (4K HDMI to MIPI CSI -2) https://toshiba.semicon-storage.com/content/dam/toshiba-ss/ncsa/en_us/docs/product-brief/assp/14A04_TC358840_ProdBrief.pdf
This is their 720p...
(https://raw.githubusercontent.com/6by9/raspiraw/master/res/Screenshot179.png)
Looks like they skip 4 and leave 4.
I would ask them why, but its so unreliable at the moment anyways, exposure issues, dcraw issues with white balance, no way to control shutter speed by simpl switches.
It needs a lot of work, and considering that it was started in 2015, i dont have high hopes.
I wonder how 2x binning looks like and whats the max res at 24fps but still , controlling exposure is tricky, there is gain instead of iso and its controlled by command line so yo ucant change stuff mid recording.
I can only imagine this is going to lead to some incredible experiments.
Magic Lantern port? :P
Quote from: Frayfray on May 02, 2020, 04:14:51 PM
There are other options toshiba TC358840 Camera Serial Interface Converter Chipset (4K HDMI to MIPI CSI -2) https://toshiba.semicon-storage.com/content/dam/toshiba-ss/ncsa/en_us/docs/product-brief/assp/14A04_TC358840_ProdBrief.pdf
That chip could be used to feed the video signal from a DSLR into a Raspberry Pi. Not sure why you'd want to do this.
Quote from: 2blackbar on May 02, 2020, 07:55:42 PM
This is their 720p...
Looks like they skip 4 and leave 4.
FYI, in 720p50/60, your Canon DSLR records one line and skips 4 (5x3 column binning / line skipping).
Exceptions:
- 5D3 (with full 5x3 binning, rather than line skipping)
- crop_rec with custom binning factors (3x3, 1x3 etc)
Quote from: 2blackbar on May 02, 2020, 07:55:42 PM
I would ask them why, but its so unreliable at the moment anyways, exposure issues, dcraw issues with white balance, no way to control shutter speed by simpl switches.
It needs a lot of work, and considering that it was started in 2015, i dont have high hopes.
I wonder how 2x binning looks like and whats the max res at 24fps but still , controlling exposure is tricky, there is gain instead of iso and its controlled by command line so yo ucant change stuff mid recording.
Well, yes. The question was whether the RPi can capture raw images. There is a proof of concept for that, and a rather good one in my opinion. Yes, it's not polished for serious use, but it's functional, and - for the intended audience of RPi, i.e. developers, hardware tinkerers, tech students - it's probably enough to start the ball rolling.
To repeat this important point:
Quote from: a1ex on May 01, 2020, 05:59:44 PM
Not for replacing DSLR cameras
BTW, the image you are looking at was taken with the *previous* generation of RPi cameras. Compared to that, the newly announced camera is a noticeable improvement, in my opinion (but don't forget I'm terrible at pixel peeping).
https://twitter.com/hashtag/ShotOnRaspberryPi
And you are no longer stuck with the default optics; you can use any other C/CS-mount lenses (heck, there were EOS M users adapting such lenses for filming), or - with adapters - pretty much any DSLR lenses.
Quote from: jpegmasterjesse on May 04, 2020, 04:28:25 AM
I can only imagine this is going to lead to some incredible experiments.
Yep.
I wonder how it compares with something like OpenMV H7plus (https://openmv.io/products/openmv-cam-h7-plus) (2592x1944 15fps, 1080p30, 720p60, 3.6x2.7mm active area, M12 mount) for real-time machine vision applications (visual servoing). In particular, I happen to be looking for a low-cost vision platform for CNC machining, where I'd like to get about 0.01mm positioning accuracy, or even micron levels if possible (with subpixel processing, of course), but the machine is relatively slow, so 3-4K resolution at 10-15 FPS, with optional cropping or pixel-binning for a higher frame rate, might be adequate. A "classic" 640x480 industrial camera is not going to do the trick. Are there other alternatives in this price range?
I'd get them both (RPi cam and OpenMV) for a comparison, but my budget is a little tight atm.
I could also consider an EOS M, but...
1) Good luck running computer vision algorithms on its general-purpose CPU (as Canon's image processing hardware is a big mystery even for us)
2) Exchanging the result with motor controllers or other hardware, is also tricky. UART to Arduino-like? USB to RPi? In any case, a second microcontroller board is needed.
Or, I could offload the processing to a RPi Zero, via USB, but I expect very large lags and very low resolution; maybe OK for a webcam, but not practical for visual servoing.
Quote from: jpegmasterjesse on May 04, 2020, 04:28:25 AM
Magic Lantern port? :P
If this camera proves interesting enough, it might happen sooner or later. Don't look at me, but at the target audience of RPi :)
Edit: some arguments against the RPi cam:
https://twitter.com/marcan42/status/1088472549715918848
Quote from: a1ex on May 04, 2020, 08:10:16 AM
That chip could be used to feed the video signal from a DSLR into a Raspberry Pi. Not sure why you'd want to do this.
maybe i misunderstood i thought the post above me was talking about 1080 or better res. and yeah i know that it is hdmi input i did say there are other options, but if you can capture a 4k 30/60 fps for $200- 250 from many new dslr or mirrorless on cheaper media for long recordings that would be great . i did reach out to the guy who wrote the raw script and maybe if possible to have a working model on pi4 and thank you for posting those links cause this has been on my mind for a while, i was not even sure that you could record raw on ras. pi, there was a board put out by orange pi RK3399 but i think that it was using older chip ,tc358749bg so no 4k,but i found this board from uk B112 4kp60 HDMI to CSI-2 bridge https://auvidea.eu/product/70513/ 4k/30 4:2:2 @ 10 bit wow i hope that people start experimenting soon not you alex i know that your very busy and i appreciate what you done for our cameras
Quote
I wonder how it compares with something like OpenMV H7plus (https://openmv.io/products/openmv-cam-h7-plus) (2592x1944 15fps, 1080p30, 720p60, 3.6x2.7mm active area, M12 mount) for real-time machine vision applications (visual servoing).
I think that's the resolution for still photography, no? Video capture seems to be 640x480px (at 60 FPS) (https://hshop.vn/products/mach-openmv-cam-h7-machine-vision-camera-module) (couldn't find in the official documentation - didn't try hard to be fair). It has Raw output though:
https://www.youtube.com/watch?v=8FVoSF34zNM
Quote from: a1ex on May 04, 2020, 08:10:16 AM
And you are no longer stuck with the default optics; you can use any other C/CS-mount lenses (heck, there were EOS M users adapting such lenses for filming), or - with adapters - pretty much any DSLR lenses.
Yeah, that's something awesome about these cameras. You could put a 12mm f/1.8 for less noise in autonomous driving applications, for example.
For anyone interested, we are building a 10K image dataset for autonomous driving that will be available for free to everybody and will be used in OpenPilot/Comma.ai:
https://github.com/commaai/comma10k
Quote from: Frayfray on May 04, 2020, 11:34:55 AM
yeah i know that it is hdmi input i did say there are other options, but if you can capture a 4k 30/60 fps for $200- 250 from many new dslr or mirrorless on cheaper media for long recordings that would be great .
Ah, implementing a custom HDMI recorder. Aren't there already good options on the market? (no experience with any of them, but I had the impression this market is already saturated).
Assuming the RPi has the processing power to handle a 4K stream (the more powerful ones might have, but I have no experience with them), it could make sense. Found these things, apparently doing exactly what you mentioned:
https://eu.mouser.com/_/?P=1y9f2kt (B110 using the TC358840)
https://auvidea.eu/product-category/csi2bridge/hdmi2csi/
https://eu.mouser.com/datasheet/2/864/B10x_technical_reference_1.4-1130369.pdf (B101 and B102)
https://auvidea.eu/download/manual/JN30/JN30_technical_reference_1.0.pdf (mentions the B112)
https://www.raspberrypi.org/forums/viewtopic.php?t=216903
https://www.youtube.com/watch?v=-r0ZN7LeSlM (https://www.youtube.com/watch?v=-r0ZN7LeSlM)
Quote
there was a board put out by orange pi RK3399
Right, this one has HDMI IN (specs unknown): http://www.orangepi.org/Orange%20Pi%20RK3399/
Quote from: Luther on May 04, 2020, 02:59:29 PM
I think that's the resolution for still photography, no? Video capture seems to be 640x480px (at 60 FPS) (https://hshop.vn/products/mach-openmv-cam-h7-machine-vision-camera-module) (couldn't find in the official documentation - didn't try hard to be fair). It has Raw output though:
https://www.youtube.com/watch?v=8FVoSF34zNM (https://www.youtube.com/watch?v=8FVoSF34zNM)
The H7 (https://openmv.io/products/openmv-cam-h7) is, indeed, limited to 640x480. The H7plus (https://openmv.io/products/openmv-cam-h7-plus) has OV5640, which is capable of 2592x1944 at 15 FPS. I don't expect this board to process such resolutions in real-time, at this frame rate, but working on cropped areas (regions of interest) might be doable.
Both have the option of a 640x480 global shutter module, or a FLIR Lepton module (thermal camera).
Quote
For anyone interested, we are building a 10K image dataset for autonomous driving that will be available for free to everybody and will be used in OpenPilot/Comma.ai:
https://github.com/commaai/comma10k
Very nice! Isn't that the self-driving car startup of George Hotz?
Quote from: a1ex on May 04, 2020, 03:37:30 PM
The H7 (https://openmv.io/products/openmv-cam-h7) is, indeed, limited to 640x480. The H7plus (https://openmv.io/products/openmv-cam-h7-plus) has OV5640, which is capable of 2592x1944 at 15 FPS. I don't expect this board to process such resolutions in real-time, at this frame rate, but working on cropped areas (regions of interest) might be doable.
Both have the option of a 640x480 global shutter module, or a FLIR Lepton module (thermal camera).
I see, I mixed up the Plus with the normal one. That's a very impressive resolution for such a small/cheap camera.
Quote
Very nice! Isn't that the self-driving car startup of George Hotz?
Yes, exactly. The project is all open source, couldn't believe it when I first saw. And it works very well most of the time.
here is a video reviewing this camera link 8)
https://www.youtube.com/watch?v=_Oh7sITnwMU
That's so awesome @Frayfray ! Very good quality. This would be very useful for autonomous driving projects or drone projects (maybe port it to smaccmpilot (https://smaccmpilot.org/hardware/iris.html)).
i would be interested in drone applications but with all the other projects going on right now they have to wait a little ,i checked with local store already sold out on this camera ! i think lots of good things are going to happen once people get there hands on few of these and to mount them in side a car is a very good idea ... :)
If if it's the Raspberry Pi VideoCore IV gpu that is hindering the capabilities of the Sony IMX477 maybe another board will do the job.
The Nvidia Jetson Nano paired with the Camera Module v2 (IMX219-200) can achieve 3840x1848 at 28 fps if I'm not mistaken.
I would stick a Kern Paillard 36mm f/2.8 D mount lens on that thing.
I haven't done much research tbh.
jestson nano cam faq
Quote1.14 Is it possible to get RAW Bayer frames from the camera?
Current camera driver and firmware doesn't support getting RAW Bayer frames from camera. We can customize the camera driver and the firmware according to your requirement. Kindly write your requirements to [email protected]. They will help you out on customization request.
I thought some may be interested in this post https://www.strollswithmydog.com/open-raspberry-pi-high-quality-camera-raw/
Quote from: 2blackbar on May 15, 2020, 11:49:33 AM
jestson nano cam faq
This is from e-consystems and concerns their camera (e-CAM55_CUMI0521_MOD an AR0521 sensor).
The Jetson nano works with the v2 IMX219 sensor out of the box.
https://leopardimaging.com/product/csi-2-mipi-modules-i-pex/csi-2-mipi-modules/li-imx477-mipi-cs/
It's 4 times the price and this is pretty much the same as the hq camera. It works on a old Jetson nano tx1 https://elinux.org/Jetson/Cameras
The driver is not open source and you'd probably need their API unfortunately (but CUDA tho plus it has 2 interfaces and you can hook up 6 with an adapter).
So maybe it would works with ease with the 50 bucks hq camera.
edit: leopard imaging's camera works with a tx2 too.
People are working on upstreaming the Broadcom Unicam CSI-2 and ISP peripheral drivers to support libcamera, that will include the hq camera.
Quote from: garry23 on May 15, 2020, 04:16:37 PM
I thought some may be interested in this post https://www.strollswithmydog.com/open-raspberry-pi-high-quality-camera-raw/
That's nuts. Such a small camera, getting these results.
https://www.strollswithmydog.com/pi-hq-cam-sensor-performance/
Technical specs are fine and dandy but so far nobody posted any decent looking raw video from these in at least HD resolution.
There are some CCTV cameras that use dual exposure that are more exciting for me achieving over 14 stops easily but they only sell with subscriptions when i asked half year ago.
I have one that I'm currently putting inside an old 8mm body - hoping to have it ready by the end of the week, so I can post some footage when it's all hooked up.
I could upload some footage already if you like, but it won't be anything more than a tripod shot filming a leaf or something... Also my knowledge of coding etc. is very limited. This is my first Pi project, so I don't know how useful I can be other than sharing footage.
Great, please record in 24fps in highest resolution you can achieve, on infinity focus so we could check aliasing, so maybe some bricks and roofs and patterns from distance.
Also If You can upload some DNG frames , can be even one from 24fps clip.
(https://www.raspberrypi.org/app/uploads/2020/04/stupid3-500x375.jpg)
[/quote]
Super cool this new camera I think the same as other colleagues, it will not be a device that will replace the DSLR but it looks very tempting to buy it, the technological future is getting better.
A few days ago I saw a film (http://www.cinecalidad.plus) that was very futuristic in the technological field and I think we are going there, of course after we leave the covid-19
Sorry it took a while, here's two test shots out my window. I'm really not sure how helpful these are as the only c-mount lens I have has sever blurring and CA especially on the edges. It's 25mm 1.4 which is very tele on this sensor. I tried it at a number of f-stops to hopefully give some variety in the detail. The issues mentioned above get worse the more you open up, and the more you close down the more ISO has to work, so I'm not sure what the happy medium is just yet. Hopefully there's at least 1 frame in here that can be of some use.
Anyway, here's the original files, which come out .264 and I also converted them to h265.mkv for easier viewing.
1920x1080
24p
mixed aperture f1.4 - f8(ish)
auto shutter speed
auto ISO
auto WB
https://www.dropbox.com/sh/seddezwwymwnh4v/AAAU5es-RCWOXCj1Fpkbz3Dba?dl=0
I'm planning to use this with D-mount lenses, but until I've finished my build, I have no way to mount them to test with nicer glass.
Thanks, looks fine, can You output DNG files, raw files from it ?
Premiere won't accept the .264 files, and I feel like taking a still from a converted file defeats the point right?
The originals are there, so feel free to take your own frames and stills.
I've not looked into taking a photo on the pi, but I believe they'd be jpeg.
Im not sure You understand, we are excited about these sensors and cameras because of RAW capability, phones can record h.264 since over a decade, its very easy to rip out original lens and glue back C mount to every smartphone.
So You cant get RAW DNG frames from the camera ? Only h.264 ? Those are subsampled at 4:2:0 , nowhere near at what RAW is.
Yes I get that, and I'm aware of how h264 is used. I highly doubt the camera will record DNG as a video stream, as there is a patent out stopping that from being possible. If you could record a DNG video stream, the biggest part of Magic Lantern, raw video, is already taken care of right?
The documentation says:
Picture formats JPEG (accelerated), JPEG + RAW, GIF, BMP, PNG, YUV420, RGB888
Video formats raw h.264 (accelerated)
10-bit RAW RGB data Yes - format conversions available via GPU
I'm unclear as to what 'raw h.264' really means, but since it's not in a wrapper (mp4 / mov) I took that to mean it's as unprocessed as you can get. Would love to be proven wrong though.
Happy to do more testing, but I'm far from a coder, so let me know what to type into terminal and I'll report what happens. (No blowup codes please)
EDIT:
I've tried using some YUV commands which, as far as my understanding goes, should bypass any conversion. I'm unable to look at either a still or a video using this command though. I've added them to the Dropbox folder... again, if they're not helpful, let me know what else I can try.
How can h.264 ever be raw :P.
What does it mean? Some format available but how?
Quote10-bit RAW RGB data Yes - format conversions available via GPU
Edit: How to decode the h.264 files? Tried viewing but fails.
Quote from: Danne on May 20, 2020, 02:29:03 PM
How can h.264 ever be raw :P.
Hi444PP profile does predictive losless coding. Don't know if it applies here.
Quote from: Danne on May 20, 2020, 02:29:03 PM
Edit: How to decode the h.264 files? Tried viewing but fails.
That's why I converted them to h265 :)
If you didn't see me edit above, there's a few more files in Dropbox.
I've also tried out a few 4x3 resolutions, since that's what I'll be shooting. The largest I could go with that is 1600x1200 - which is fairly reasonable. However I did read previously I should be able to get up to 30fps at 2028x1520, which would be really nice... so far it just gives me errors with that though.
we have to get to that raw 10 bit rgb data stream and debayer it , maybe try some raw to dng converters
Digital super 8 eh? id do it too with this setup
I've tried a few, like SlimRaw, but they're not recognising the file. I feel like it needs an extension, but it's anybodies guess as to what extension it needs. I'll keep reading about it and offer more footage as and when more info becomes available.
Quote from: 2blackbar on May 20, 2020, 06:07:24 PM
Digital super 8 eh? id do it too with this setup
Yeah, the sensor size is almost exactly the same as 8mm. and the Pi fits very nicely inside my (previously broken and unusable) Bolex B8.
EDIT:
You can playback the native pi files (that end in .264) in VLC - but I've not found any other app that can open them.
Interestingly, the app Scoop! has this to say about those files:
200520075652.h264
Format : AVC
Format/Info : Advanced Video Codec
File size : 5.10 MiB
Video
Format : AVC
Format/Info : Advanced Video Codec
Format profile : High@L4
Format settings : CABAC / 1 Ref Frames
Format settings, CABAC : Yes
Format settings, ReFrames : 1 frame
Width : 1 920 pixels
Height : 1 080 pixels
Display aspect ratio : 16:9
Color space : YUV
Chroma subsampling : 4:2:0
Bit depth : 8 bits
Scan type : Progressive
I don't know if any of that is useful to anyone, but it certainly can't hurt to post it.
yeah post all findings, as you see its 4:2:0, typical smartphone vid specs
@Anaconda_ , .h264 is a basic avc/mp4/mov without the wrapper of mp4/mov .
It's consider raw AVC , because of no wrapper . I've use this format many times in windows.
https://extension.nirsoft.net/h264
That profile you posted is the same profile canon 5d3/d5 cams .mov/avc use .
The software a1ex posted before, raspiraw, added initial support for this camera's sensor:
https://github.com/6by9/raspiraw
https://github.com/6by9/raspiraw/commit/dbe1acf64cba221787080ad06f79d3a5bccd171a
Technically, it would be possible to record raw 24fps. RPi 4 has USB 3.0 ports, you could attach a SDD using an adapter.
They could add support for MLV format in rapiraw using Ilia's lib, so we could process using MLVApp... just an idea :P
In theory, the files 'rawvid' and 'rawtest' in the Dropbox link should be raw video, not processed or converted into anything... but I have no idea how to go about looking at them, or looking at whatever information they might hold.
( https://www.dropbox.com/sh/seddezwwymwnh4v/AAAU5es-RCWOXCj1Fpkbz3Dba?dl=0 )
With my limited knowledge, Raspiraw helps decode this and then another program (dcraw) wraps it in a ppm container. I understand the concept and everything in theory, but actually getting it to work in practice is beyond my comprehension. In the words of the developer:
'If you expect everything spoon-fed to you, then raspiraw is not the program for you.'
I love ML on my Canon cameras. Thanks to the efforts of the people behind it, it's easy to understand, set up and work with. Raspiraw seems to be very different from a user's perspective... maybe the backend of ML is similar, but it's hidden behind the UI - I have no idea.
Not wanting to sound selfish, but my camera setup is working as is, albeit with .264 outputs... My switches run the codes I need them to, and I very much hope that in the future I can record some sort of raw, be it DNG, or MLV.
I'd love to be able to help out here, but without being instructed to try specific things, I don't know what I can do. I'll continue to do my own testing and post results, so please feel free to recommend certain (and specific) things, but keep in mind I am not a programmer and beyond the basics, learning python is a pain in my ass.
With all of that said, I do have a VNC on my pi and it's setup to work with remote connections, so I'm happy to connect with people and let them control my system and test scripts and stuff to give someone with more knowledge a chance to make some progress...
I've added a .rgb file, which should be fairly raw. I've read you can decode it with dcraw, I have dcraw installed, and am just working out how to use it.
I'm not having any luck and need to stop working on this project for now.. Anyone else can feel free to try and get things working :)
Let me know how you get on!
Was about time somebody got around to this right? ;)
Sample footage shot from the Raspberry Pi HQ Camera, 12-bit DNG 2028x0180 @ 24fps
You can read more here: https://www.raspberrypi.org/forums/viewtopic.php?f=43&t=296776
If you can/want to help contribute, stay tuned I'll be posting guides/source to how I did it. Hopefully it can grow to a working camera system eventually.
I found this raspberry pi camera(UYVY): https://www.e-consystems.com/raspberry-pi-4/high-quality-mipi-csi-camera.asp
What's your opinion about this camera?
Can someone correct me if I am wrong, but is this not the exact same concept as Axiom Camera? Open Source, all schematics on the table, if you got the skill you change any part you see fit? Or is there a fundamental difference, if so, what is it?
I must admit I personally only know rasperry pi as a TV computer.
Kharak its "just" a sensor with c mount
forget it, all pi card writing speeda are crap even if sensor outputs raw its a waste pf time until they upgrade components but its gonna be awhile , got to milk these old components first.
Use phone, its just a funky webcam sold separstely
raw in 1080 is fne i guess but card readers on pi are bootlenecking the process, there wont be brewkthrough witjout faster ones but id love to be proven wrong of coures , read specs and its all clear.
Yellowguy whats the highest resolution in 24fpa this sony sensor can output to usb3.0 ssd? theory i mean.
the idea of working on rapberry softwareto enable raw recording is great but current specs of components isnt good enough for above 2k
Is it 1080 skipped or binned ? If its possible to get 4k raw with current components on raspi biards , ill buy one or two.
You should kickstart it, fundraise it.
In order to get 24fps@4056 (H) × 3040 (V) with 12bit RAW, the total bandwidth requires 4056x3040x24fpsx12/8=443.89MB/s
--
Soooo... i did some digging and usb 3.0 on raspberry 4 (best newest one) is able to push 620 megabits per second at max...
620 Megabit/Second (Mbit/s) = 77.5 Megabyte/Second (MB/s)
Which means its not enough to intercept 4K frames at 24fps , maybe if raw stream would be compressed but that would reduce fps unless done by hardware.
OR is it 5 Gbps which means 625 MB/s ? Cause thats definitely capable of 4k raw acquisition.
OK some guys are able to get around 340MB per second with SSD drives on pi 4
So thats barely enough for 4k raw stream.
2:30:
So... sigma FP camera outputs 12 bit raw frame at 2400Mbps which is 300MB per sec data rate , there might be a chance , depending how heave one dng 4k frame from pi is.
So i might be proven wrong eventually which is amazing.
Thanks for clarifying.
I borrowed a Pi once from a friend, but it was only for watching films on the TV.
Thats just assumptions based on specs, until its done then nothing is sure, IMX477R sony sensor might have better chance to output raw 4k in 24fps but raspi 4 wont be able to spit it out to SSD fast enough according to some users , depending on whos talking theres various info about it.
But still, its only raw video feed, no mic.Probably no preview at all as that would drop fps.
Hi...FWIW my first catch with the HQ camera (utilizing raspistill) seemed as though irregular pixel clamor, yet for certain edges from the real scene scarcely apparent. I was utilizing it's anything but a more established Pi that hadn't been refreshed in some time, and refreshing it (sudo able update; sudo well-suited redesign) fixed the issue. I haven't heard any other person announcing that, so I get it is uncommon.