NASA Mars Panos: Glaring Stitching Errors!!??!!

Discussion in 'Digital Photography' started by brian, Jan 5, 2004.

  1. brian

    brian Guest

    brian, Jan 5, 2004
    #1
    1. Advertising

  2. brian

    Mark Herring Guest

    On 5 Jan 2004 12:05:28 -0800, (brian) wrote:

    >When will NASA get wind of Panorama Tools so they can do things
    >right?? http://marsrovers.jpl.nasa.gov/gallery/press/spirit/20040105a/PIA04991.jpg
    >
    >Also, the world's most expensive panoramic camera doesn't appear to
    >have the ability to keep the entrance pupil of the lens fixed while
    >making pitch and yaw rotations. Groan.
    >
    >Brian
    >www.caldwellphotographic.com


    These cameras are not intended to produce hi-quality panos. When
    panos are made, it is all downstream in the processing---you'll see
    better ones as the mission progresses.

    BTW, the constraint often qouted --ie rotate around the front nodal
    point (pupil)---is only im,portant if there is a large distance range.
    I do panos in Panotools / PT assembler---it corrects all errors and I
    just shoot naturally. (Good SW makes lazy photographer?)
    **************************
    Mark Herring, Pasadena, Calif.
    Private e-mail: Just say no to "No".
    Mark Herring, Jan 5, 2004
    #2
    1. Advertising

  3. brian

    Annika1980 Guest

    Annika1980, Jan 5, 2004
    #3
  4. brian

    DJ Guest

    DJ, Jan 5, 2004
    #4
  5. brian

    Ron Hunter Guest

    brian wrote:

    > When will NASA get wind of Panorama Tools so they can do things
    > right?? http://marsrovers.jpl.nasa.gov/gallery/press/spirit/20040105a/PIA04991.jpg
    >
    > Also, the world's most expensive panoramic camera doesn't appear to
    > have the ability to keep the entrance pupil of the lens fixed while
    > making pitch and yaw rotations. Groan.
    >
    > Brian
    > www.caldwellphotographic.com


    The camera used really isn't the one intended for general use, just a
    'spotting scope', as it were. expect better pictures later.
    Ron Hunter, Jan 5, 2004
    #5
  6. brian

    Tumbleweed Guest

    "DJ" <> wrote in message
    news:...
    > On 05 Jan 2004 21:22:44 GMT, (Annika1980) wrote:
    >
    > >>From: (brian)

    > >
    > >>When will NASA get wind of Panorama Tools so they can do things
    > >>right??

    >
    >>http://marsrovers.jpl.nasa.gov/gallery/press/spirit/20040105a/PIA04991.jpg
    > >
    > >Maybe they just used the software that came with the Rover?

    >
    > LOL!


    And they'd better hurry up, I understand it expires after 30 days unless
    they pay for the full version.

    --
    Tumbleweed

    Remove theobvious before replying (but no email reply necessary to
    newsgroups)
    Tumbleweed, Jan 5, 2004
    #6
  7. brian

    DJ Guest

    On Mon, 5 Jan 2004 22:04:46 -0000, "Tumbleweed"
    <> wrote:

    >"DJ" <> wrote in message
    >news:...
    >> On 05 Jan 2004 21:22:44 GMT, (Annika1980) wrote:
    >>
    >> >>From: (brian)
    >> >
    >> >>When will NASA get wind of Panorama Tools so they can do things
    >> >>right??

    >>
    >>>http://marsrovers.jpl.nasa.gov/gallery/press/spirit/20040105a/PIA04991.jpg
    >> >
    >> >Maybe they just used the software that came with the Rover?

    >>
    >> LOL!

    >
    >And they'd better hurry up, I understand it expires after 30 days unless
    >they pay for the full version.


    Then after 6 months it will start running intolerably slowly and requiring
    frequent re-boots, at which stage NASA will have to upgrade to Rover 05.
    DJ, Jan 5, 2004
    #7
  8. brian

    Annika1980 Guest

    >From: "Tumbleweed"

    >> >Maybe they just used the software that came with the Rover?

    >>
    >> LOL!

    >
    >And they'd better hurry up, I understand it expires after 30 days unless
    >they pay for the full version.


    And if it's Adobe, ET will have to phone home to activate the software.
    Annika1980, Jan 5, 2004
    #8
  9. brian

    Nik Simpson Guest

    DJ wrote:
    > On Mon, 5 Jan 2004 22:04:46 -0000, "Tumbleweed"
    > <> wrote:
    >
    >> "DJ" <> wrote in message
    >> news:...
    >>> On 05 Jan 2004 21:22:44 GMT, (Annika1980) wrote:
    >>>
    >>>>> From: (brian)
    >>>>
    >>>>> When will NASA get wind of Panorama Tools so they can do things
    >>>>> right??
    >>>
    >>>> http://marsrovers.jpl.nasa.gov/gallery/press/spirit/20040105a/PIA04991.jpg
    >>>>
    >>>> Maybe they just used the software that came with the Rover?
    >>>
    >>> LOL!

    >>
    >> And they'd better hurry up, I understand it expires after 30 days
    >> unless they pay for the full version.

    >
    > Then after 6 months it will start running intolerably slowly and
    > requiring frequent re-boots, at which stage NASA will have to upgrade
    > to Rover 05.


    Not as silly as it sounds, according to the Nova program about it today,
    they did a software upgrade to the Rover while it was on the way to Mars!


    --
    Nik Simpson
    Nik Simpson, Jan 6, 2004
    #9
  10. brian

    brian Guest

    Mark Herring <> wrote in message news:<>...
    > On 5 Jan 2004 12:05:28 -0800, (brian) wrote:
    >
    > >When will NASA get wind of Panorama Tools so they can do things
    > >right?? http://marsrovers.jpl.nasa.gov/gallery/press/spirit/20040105a/PIA04991.jpg
    > >
    > >Also, the world's most expensive panoramic camera doesn't appear to
    > >have the ability to keep the entrance pupil of the lens fixed while
    > >making pitch and yaw rotations. Groan.
    > >
    > >Brian
    > >www.caldwellphotographic.com

    >
    > These cameras are not intended to produce hi-quality panos. When
    > panos are made, it is all downstream in the processing---you'll see
    > better ones as the mission progresses.
    >
    > BTW, the constraint often qouted --ie rotate around the front nodal
    > point (pupil)---is only im,portant if there is a large distance range.
    > I do panos in Panotools / PT assembler---it corrects all errors and I
    > just shoot naturally. (Good SW makes lazy photographer?)
    > **************************
    > Mark Herring, Pasadena, Calif.
    > Private e-mail: Just say no to "No".


    Its the entrance pupil, not the nodal point. Try a stop shift
    experiment if you don't believe it. Sure, if everything in the image
    is far away then keeping the entrance pupil stationary isn't
    important. But this does not appear to be the case for the Mars
    rover, and its rarely the case for interesting photography on earth.
    I realize that stereo photography is incompatible with keeping the
    pupil stationary, but I would have designed the system so that at
    least one of the cameras could take a flawless high resolution
    stitched panorama.

    Brian
    www.caldwellphotographic.com
    brian, Jan 6, 2004
    #10
  11. brian

    Guest

    (brian) wrote:

    > Also, the world's most expensive panoramic camera doesn't appear to
    > have the ability to keep the entrance pupil of the lens fixed while
    > making pitch and yaw rotations. Groan.


    I thought all panorama software estimated the projections of the
    original images and then re-projected them onto the larger canvas.
    Wouldn't it have to? There is no way Random J. Photographer with a
    handheld camera is going to be able to maintain an accurate entrance
    pupil while snapping off the source images.

    The JPL images look like they were assembled by hand, in a hurry, with
    material they had available (some of it of questionable quality).
    Re-shoots aren't easy when the pipe is a ~hundred million kilometres
    long and tens(?) of kilobits/second wide.
    , Jan 6, 2004
    #11
  12. (brian) writes:

    > Mark Herring <> wrote in message news:<>...
    > > On 5 Jan 2004 12:05:28 -0800, (brian) wrote:
    > >
    > > >When will NASA get wind of Panorama Tools so they can do things
    > > >right?? http://marsrovers.jpl.nasa.gov/gallery/press/spirit/20040105a/PIA04991.jpg
    > > >
    > > >Also, the world's most expensive panoramic camera doesn't appear to
    > > >have the ability to keep the entrance pupil of the lens fixed while
    > > >making pitch and yaw rotations. Groan.


    Perhaps because it's a pair of cameras for stereo. It's hard to keep both
    camera viewpoints fixed and still make a full-circle panorama :).

    <snip>

    --
    -Stephen H. Westin
    Any information or opinions in this message are mine: they do not
    represent the position of Cornell University or any of its sponsors.
    Stephen H. Westin, Jan 6, 2004
    #12
  13. brian

    kpfeif Guest

    wrote in message news:<>...
    > (brian) wrote:
    >
    > > Also, the world's most expensive panoramic camera doesn't appear to
    > > have the ability to keep the entrance pupil of the lens fixed while
    > > making pitch and yaw rotations. Groan.

    >
    > I thought all panorama software estimated the projections of the
    > original images and then re-projected them onto the larger canvas.
    > Wouldn't it have to? There is no way Random J. Photographer with a
    > handheld camera is going to be able to maintain an accurate entrance
    > pupil while snapping off the source images.
    >
    > The JPL images look like they were assembled by hand, in a hurry, with
    > material they had available (some of it of questionable quality).
    > Re-shoots aren't easy when the pipe is a ~hundred million kilometres
    > long and tens(?) of kilobits/second wide.



    Go ahead and download Maestro at http://mars.telascience.org/

    You'll be able to manipulate the raw images provided by the rovers.
    It'll also create anaglyphs for you to take a gander at...using those
    50s-like 3d glasses.
    kpfeif, Jan 6, 2004
    #13
  14. brian

    brian Guest

    wrote in message news:<>...
    > (brian) wrote:
    >
    > > Also, the world's most expensive panoramic camera doesn't appear to
    > > have the ability to keep the entrance pupil of the lens fixed while
    > > making pitch and yaw rotations. Groan.

    >
    > I thought all panorama software estimated the projections of the
    > original images and then re-projected them onto the larger canvas.
    > Wouldn't it have to? There is no way Random J. Photographer with a
    > handheld camera is going to be able to maintain an accurate entrance
    > pupil while snapping off the source images.
    >
    > The JPL images look like they were assembled by hand, in a hurry, with
    > material they had available (some of it of questionable quality).
    > Re-shoots aren't easy when the pipe is a ~hundred million kilometres
    > long and tens(?) of kilobits/second wide.



    You've got the basic algorithm correct, and with proper technique the
    stitching errors can be brought to less than 1 pixel even for
    extremely large mosaics. Of course, proper technique generally
    requires a tripod and pano head unless everything is effectively at
    infinity. Hopefully you are right about the preliminary nature of the
    images, and future ones will be taken and processed correctly.

    Brian
    www.caldwellphotographic.com
    brian, Jan 7, 2004
    #14
  15. brian

    Ron Hunter Guest

    wrote:

    > (brian) wrote:
    >
    >
    >>Also, the world's most expensive panoramic camera doesn't appear to
    >>have the ability to keep the entrance pupil of the lens fixed while
    >>making pitch and yaw rotations. Groan.

    >
    >
    > I thought all panorama software estimated the projections of the
    > original images and then re-projected them onto the larger canvas.
    > Wouldn't it have to? There is no way Random J. Photographer with a
    > handheld camera is going to be able to maintain an accurate entrance
    > pupil while snapping off the source images.
    >
    > The JPL images look like they were assembled by hand, in a hurry, with
    > material they had available (some of it of questionable quality).
    > Re-shoots aren't easy when the pipe is a ~hundred million kilometres
    > long and tens(?) of kilobits/second wide.


    They look like manual pasteups to me. I suspect that they are in quite
    a rush to get them out. I am SURE they have software that can do a
    better job of stiching that what I have seen so far...
    Ron Hunter, Jan 7, 2004
    #15
  16. brian

    Ron Hunter Guest

    kpfeif wrote:

    > wrote in message news:<>...
    >
    >> (brian) wrote:
    >>
    >>
    >>>Also, the world's most expensive panoramic camera doesn't appear to
    >>>have the ability to keep the entrance pupil of the lens fixed while
    >>>making pitch and yaw rotations. Groan.

    >>
    >>I thought all panorama software estimated the projections of the
    >>original images and then re-projected them onto the larger canvas.
    >>Wouldn't it have to? There is no way Random J. Photographer with a
    >>handheld camera is going to be able to maintain an accurate entrance
    >>pupil while snapping off the source images.
    >>
    >>The JPL images look like they were assembled by hand, in a hurry, with
    >>material they had available (some of it of questionable quality).
    >>Re-shoots aren't easy when the pipe is a ~hundred million kilometres
    >>long and tens(?) of kilobits/second wide.

    >
    >
    >
    > Go ahead and download Maestro at http://mars.telascience.org/
    >
    > You'll be able to manipulate the raw images provided by the rovers.
    > It'll also create anaglyphs for you to take a gander at...using those
    > 50s-like 3d glasses.


    I did, but where do I go for the raw images?
    Ron Hunter, Jan 7, 2004
    #16
  17. In article <>,
    (kpfeif) wrote:

    > Go ahead and download Maestro at http://mars.telascience.org/


    > You'll be able to manipulate the raw images provided by the rovers.
    > It'll also create anaglyphs for you to take a gander at...using those
    > 50s-like 3d glasses.



    I've tried Maestro in "tutorial" mode.
    On my computer the rover has large blank areas under it and there is
    terrain on top of the solar panels.
    How does it look on yours?
    I'm wondering if it is their problem or mine.
    --
    Charlie Dilks
    Newark, DE USA
    Charlie Dilks, Jan 7, 2004
    #17
    1. Advertising

Want to reply to this thread or ask your own question?

It takes just 2 minutes to sign up (and it's free!). Just click the sign up button to choose a username and then you can ask your own questions on the forum.
Similar Threads
  1. Jeanne Medley
    Replies:
    2
    Views:
    1,074
    RHODRI REES
    Feb 11, 2004
  2. Bob

    Spherical Panos using Fisheye Attachments?

    Bob, Dec 31, 2003, in forum: Digital Photography
    Replies:
    1
    Views:
    763
    Terje Mathisen
    Jan 1, 2004
  3. takeone

    Digicam on Mars -- Spirit Mars Rover

    takeone, Jan 5, 2004, in forum: Digital Photography
    Replies:
    1
    Views:
    466
    Chris P in PA
    Jan 6, 2004
  4. Mark M

    Hmmm. Panos are kinda nifty.

    Mark M, Jul 29, 2004, in forum: Digital Photography
    Replies:
    2
    Views:
    302
    Brian C. Baird
    Jul 29, 2004
  5. Philippe

    Stitching panos

    Philippe, Aug 18, 2006, in forum: Digital Photography
    Replies:
    15
    Views:
    584
    Philippe
    Aug 21, 2006
Loading...

Share This Page