10
51.9 MP camera app

  1. Anforapps
    Donut Mar 3, 2016

    Anforapps , Mar 3, 2016 :
    Hi all,

    new version of "Camera Super Pixel" (beta) app is also compatible with OnePlus 2 (supports 51.9 megapixels pics).

    The app takes a burst of pictures and intelligently combines them into a higher resolution photograph (up to 4x) using an own-developed super-resolution (SR) algorithm. The camera is optimized for static scenes. In other situations the result may not be good.

    You can get more info about super-resolution here: https://en.wikipedia.org/wiki/Super-resolution_imaging

    We are working on improving the performance in terms of speed and quality of the result.

    It's on beta stage, please give us feedback if you have any problem. Thanks!

    https://play.google.com/store/apps/details?id=com.anforapps.camerasuperpixel
     
    Last edited: Mar 4, 2016

    #1
  2. aditlal
    Eclair Mar 3, 2016

    aditlal , Mar 3, 2016 :
    How does it replicate 51 mp shots just curious , testing app by burst shooting
     

    #2
  3. Anforapps
    Donut Mar 3, 2016

    Anforapps , Mar 3, 2016 :
    How it works: burst of photos (8 pics), alignment with subpixel accuracy and mixing best (up to 8) images.
     

    #3
    IvanMuse, otto2 and royz0r like this.
  4. Andrejwk0
    Eclair Mar 3, 2016


    #4
  5. funmunke
    Marshmallow Mar 3, 2016

    funmunke , Mar 3, 2016 :
    ....why would I want a 51mp photo? Even 25 is excessive unless you are compensating or a lack of zoom. 51 would be a massive file and wasteful for most shots.
     

    #5
    Document02, p51d007 and ogriff like this.
  6. ogriff
    Jelly Bean Mar 3, 2016

    ogriff , Mar 3, 2016 :
    Surely the lens quality on a phone compromises picture quality way beyond the merit of extra pixels?
     

    #6
  7. apisarenco
    Gingerbread Mar 3, 2016

    apisarenco , Mar 3, 2016 :
    I played with it for a bit.

    Don't kid yourself folks, it's not 50 megapixels. Not even close. It's not even 13 megapixels.

    But it does seem a tiny bit better quality than with the stock camera. Background blur is softer in low light (less noise) and very fine details are a tiny bit more detailed.

    I'd say there's 50-100% more information in the photo.
    When resized to 13 megapixels, the JPEG does get about 2 times bigger (accounted for JPEG quality). So it does get more info.

    Verdict - only useful in low light, in still images, where you'd do anything for a bit more fine detail.
    For the rest - best use the stock app with RAW support, that lets you get the dynamic range that you want, and some extra editing capabilities without losing any quality.
     

    #7
    burnout123 and otto2 like this.
  8. delpinsky
    Ice Cream Sandwich Mar 3, 2016

    delpinsky , Mar 3, 2016 :
    Nothing more nothing less than what the ColoOS ported camera app does with its HD plug-in.
    A software interpolation.

    p.s.
    I see now it's not compatible with the OnePlus One. Too bad :(
    It would have been nice to compare it directly with the ColorOS app.
     
    Last edited: Mar 3, 2016

    #8
    otto2 likes this.
  9. M4ENY
    KitKat Mar 3, 2016

    M4ENY , Mar 3, 2016 :
    so how is this different from lets a panorama shot already available on stock camera apps?
    I am just curious as to what the benefits of your app is :)
     

    #9
  10. Dunnow
    The Lab - OnePlus 6 Reviewer Mar 3, 2016

    Dunnow , Mar 3, 2016 :
    this is what I don't understand. The lenses of the camera are giving you a field of view, how can the app take 8 full rez pictures of that and create a bigger one?
     

    #10
  11. metalerjf
    Gingerbread Mar 3, 2016

    metalerjf , Mar 3, 2016 :
    hi, could you tell me where I can download the app? :D
     

    #11
  12. tweakersinv
    Eclair Mar 4, 2016


    #12
  13. p51d007
    KitKat Mar 4, 2016

    p51d007 , Mar 4, 2016 :
    LOL, nothing against the developer, but, 51mp interpolated, on a pinhole sensor?
    My dSLR takes better photos than any super megapixel smartphone and it's only 12mp.
    Why? THE GLASS AND SIZE of the sensor. Yeah, smartphones aren't "bad" for snapshots, but for photographs, you need at minimum an APS-C sensor.
     

    #13
  14. apisarenco
    Gingerbread Mar 4, 2016

    apisarenco , Mar 4, 2016 :
    all photos are from the same point, with the same angle, in the same direction. They are assembled into one, taking as much detail as possible from all of them.
    the image is resized, and details added. Frankly I think it's just a marketing ploy, since even if you resize it to 13 megapixels, it's far from being very sharp.
    that's elitist talk. Photographs were taken with extremely low quality compared to today's equipment. You can take photographs with anything, as long as your hands and mind work correctly. And it's not only me who is saying that. Watch "Pro tog, cheap camera challenge" videos. See kiddie cameras, barbie cams, and similar, be used for low quality, but quite interesting photographs.

    It's not the equipment that makes a photographer.
     

    #14
    M4ENY and otto2 like this.
  15. rjainbwr1975
    Cupcake Mar 4, 2016

  16. xoma262
    Jelly Bean Mar 4, 2016


    #16
    Gaurav Rohilla likes this.
  17. M4ENY
    KitKat Mar 4, 2016

    M4ENY , Mar 4, 2016 :
    Oo I see so it's a really detailed normal photo. Sounds interesting
     

    #17
  18. Wahoux
    Lollipop Mar 4, 2016


    #18
  19. kadiix
    Ice Cream Sandwich Mar 4, 2016


    #19
  20. Anforapps
    Donut Mar 4, 2016

    Anforapps , Mar 4, 2016 :
    Hi Dunnow, the images taken in burst have small displacements, calculate precisely these displacement allows us to reconstruct a high resolution image and reduce noise.
    You can get more info about super-resolution here: https://en.wikipedia.org/wiki/Super-resolution_imaging

    Hi apisarenco, quality of the final photo depends on the number of images (max 8) of the burst the app can mix correctly. You can check this if you enable the "Debug mode" inside the app. Anyway, we are in beta phase and we are working hard to improve the algorithm (and the speed) to align correctly the pictures. Please, ask me if you have any other question. Thanks!
     

    #20