Loading ...
Sorry, an error occurred while loading the content.

Re: [PanoToolsNG] angle of view ???

Expand Messages
  • michel thoby
    Hi Ayrton, I think the answer is about 167 degrees. As you requested it, I have tried to develop a technique to address this unusual specific problem. I have
    Message 1 of 9 , Nov 1, 2007
    • 0 Attachment
      Hi Ayrton,

      I think the answer is about 167 degrees.

      As you requested it, I have tried to develop a technique to address
      this unusual specific problem.

      I have used only one tool for this: PTGui.

      Illustration given here:http://michel.thoby.free.fr/PB_FOV_Ayrton/
      ceiling_Ayrton_167deg.jpg
      and the script is here: http://michel.thoby.free.fr/PB_FOV_Ayrton/
      Ayrton-167deg.pts

      Assuming the provided image to be a purely partial cylindrical
      panorama, I have set the Lens Settings parameters correspondingly.
      Then I have searched for feature(s) that would encompass a large
      angle on the ceiling or on the floor and that for sure is on the same
      straight line in real world.

      I have found and selected such a feature on the ceiling: the straight
      line (in real world) stretches from the top left to the top right of
      the image but is cropped out in the center.
      By moving the image around on the grid of the Panorama Editor, I have
      forced this line to lie on the center horizontal axis and tweaked the
      FOV value step by step. When I have set to 167 degrees, the "line" is
      coincident with the axis (and the other supposed straight lines of
      the ceiling are about parallel as they should be).
      Moving away from this 167 degrees value setting makes the expected
      coincidence (of the straight line with the axis) impossible...

      I am not sure that this is perfectly accurate as my assumptions may
      be wrong. But it looks good;-)

      Regards,

      Michel

      Le 1 nov. 07 à 17:21, AYRTON - avi a écrit :

      > hi Ian,
      > THANKS
      > I tought the same, (130º) looking at the ceilings grid, but I wish to
      > know if there's a tecnique to find out the exact angle of any given
      > shot ?!?
      >
      > Ayrton
      >
      > On 11/1/07, Ian Wood <panolists@...> wrote:
      > > *IF* the grid on the ceiling is a regular grid then it's somewhere
      > > around 130 degrees, but there's some weird stuff going on above the
      > > chiller cabinets...
      > >
      > > Finding out the exact FoV would be very tricky.
      > >
      > > Ian
      > >
      > > On 1 Nov 2007, at 15:31, AYRTON - avi wrote:
      > >
      > > > Guys
      > > > I have a question, and really appreciate if someone can help me :
      > > >
      > > > I just received a photo from a client as a sample, and I would
      > like to
      > > > know if there is a way to find out what is the exact horizontal
      > angle
      > > > of view on a given image ???
      > > >
      > > > the photo is here:
      > > > www.ayrton.com/diversos/questions/angleofview.jpg
      > > >
      > > > Thanks a lot
      > > > Ayrton
      > > >
      > > > --
      > > > AYRTON 21-9982.6313 www.ayrton.com
      > > > Ladeira de Nossa Senhora, 214 / sl. 101 www.vrfolio.com
      > > > Outeiro da Glória - RJ - 22211-100 - Brasil www.vr-images.com
      > > > Panoramas do Rio de Janeiro www.rio360.com.br
      > > >
      > > >
      > > > --
      > > >
      > > > Yahoo! Groups Links
      > > >
      > > >
      > > >
      > >
      > >
      > >
      > > --
      > >
      > > Yahoo! Groups Links
      > >
      > >
      > >
      > >
      >
      > --
      > AYRTON 21-9982.6313 www.ayrton.com
      > Ladeira de Nossa Senhora, 214 / sl. 101 www.vrfolio.com
      > Outeiro da Glória - RJ - 22211-100 - Brasil www.vr-images.com
      > Panoramas do Rio de Janeiro www.rio360.com.br
      >
      >



      [Non-text portions of this message have been removed]
    • AYRTON - avi
      Michel avant de tout I must say I m a admirer of your tests, website, photos and experiments :-) Second: thanks a lot for your time. I m gonna try your
      Message 2 of 9 , Nov 1, 2007
      • 0 Attachment
        Michel
        avant de tout
        I must say I'm a admirer of your tests, website, photos and experiments :-)

        Second:
        thanks a lot for your time.
        I'm gonna try your description below

        Thanks again !
        merci beaucoup :-)

        Salut
        Ayrton


        On 11/1/07, michel thoby <thobymichel@...> wrote:
        > Hi Ayrton,
        >
        > I think the answer is about 167 degrees.
        >
        > As you requested it, I have tried to develop a technique to address
        > this unusual specific problem.
        >
        > I have used only one tool for this: PTGui.
        >
        > Illustration given here:http://michel.thoby.free.fr/PB_FOV_Ayrton/
        > ceiling_Ayrton_167deg.jpg
        > and the script is here: http://michel.thoby.free.fr/PB_FOV_Ayrton/
        > Ayrton-167deg.pts
        >
        > Assuming the provided image to be a purely partial cylindrical
        > panorama, I have set the Lens Settings parameters correspondingly.
        > Then I have searched for feature(s) that would encompass a large
        > angle on the ceiling or on the floor and that for sure is on the same
        > straight line in real world.
        >
        > I have found and selected such a feature on the ceiling: the straight
        > line (in real world) stretches from the top left to the top right of
        > the image but is cropped out in the center.
        > By moving the image around on the grid of the Panorama Editor, I have
        > forced this line to lie on the center horizontal axis and tweaked the
        > FOV value step by step. When I have set to 167 degrees, the "line" is
        > coincident with the axis (and the other supposed straight lines of
        > the ceiling are about parallel as they should be).
        > Moving away from this 167 degrees value setting makes the expected
        > coincidence (of the straight line with the axis) impossible...
        >
        > I am not sure that this is perfectly accurate as my assumptions may
        > be wrong. But it looks good;-)
        >
        > Regards,
        >
        > Michel
        >
        > Le 1 nov. 07 à 17:21, AYRTON - avi a écrit :
        >
        >
        > > hi Ian,
        > > THANKS
        > > I tought the same, (130º) looking at the ceilings grid, but I wish to
        > > know if there's a tecnique to find out the exact angle of any given
        > > shot ?!?
        > >
        > > Ayrton
        > >
        > > On 11/1/07, Ian Wood <panolists@...> wrote:
        > > > *IF* the grid on the ceiling is a regular grid then it's somewhere
        > > > around 130 degrees, but there's some weird stuff going on above the
        > > > chiller cabinets...
        > > >
        > > > Finding out the exact FoV would be very tricky.
        > > >
        > > > Ian
        > > >
        > > > On 1 Nov 2007, at 15:31, AYRTON - avi wrote:
        > > >
        > > > > Guys
        > > > > I have a question, and really appreciate if someone can help me :
        > > > >
        > > > > I just received a photo from a client as a sample, and I would
        > > like to
        > > > > know if there is a way to find out what is the exact horizontal
        > > angle
        > > > > of view on a given image ???
        > > > >
        > > > > the photo is here:
        > > > > www.ayrton.com/diversos/questions/angleofview.jpg
        > > > >
        > > > > Thanks a lot
        > > > > Ayrton
        > > > >
        > > > > --
        > > > > AYRTON 21-9982.6313 www.ayrton.com
        > > > > Ladeira de Nossa Senhora, 214 / sl. 101 www.vrfolio.com
        > > > > Outeiro da Glória - RJ - 22211-100 - Brasil www.vr-images.com
        > > > > Panoramas do Rio de Janeiro www.rio360.com.br
        > > > >
        > > > >
        > > > > --
        > > > >
        > > > > Yahoo! Groups Links
        > > > >
        > > > >
        > > > >
        > > >
        > > >
        > > >
        > > > --
        > > >
        >
        > > > Yahoo! Groups Links
        > > >
        > > >
        > > >
        > > >
        > >
        > > --
        > > AYRTON 21-9982.6313 www.ayrton.com
        > > Ladeira de Nossa Senhora, 214 / sl. 101 www.vrfolio.com
        > > Outeiro da Glória - RJ - 22211-100 - Brasil www.vr-images.com
        > > Panoramas do Rio de Janeiro www.rio360.com.br
        > >
        > >
        >
        >
        >
        > [Non-text portions of this message have been removed]
        >
        >
        >
        > --
        >
        > Yahoo! Groups Links
        >
        >
        >
        >


        --
        AYRTON 21-9982.6313 www.ayrton.com
        Ladeira de Nossa Senhora, 214 / sl. 101 www.vrfolio.com
        Outeiro da Glória - RJ - 22211-100 - Brasil www.vr-images.com
        Panoramas do Rio de Janeiro www.rio360.com.br
      • Milko Amorth
        Hi Michel, ... Interesting forensic. I would have guessed over 180° fov. To me it does not look like a cylindrical projection at all. It looks more like an
        Message 3 of 9 , Nov 1, 2007
        • 0 Attachment
          Hi Michel,

          > I am not sure that this is perfectly accurate as my assumptions may
          > be wrong. But it looks good;-)

          Interesting forensic.
          I would have guessed over 180° fov. To me it does not look like a
          cylindrical projection at all. It looks more like an array of horizontal
          rectlinear tiles fudged together, rather unsuccessfully in the ceiling
          area. Again, pure speculation, but fun.

          Cheers, Milko



          --
          Milko K. Amorth
          360° Immersive Imaging Productions

          VRCanada.ca
          Vancouver.360cities.net
          Ph:604.561.5101
          fx: 604.909.5125

          Skype: VRdundee
          Adobe Photographers Directory: tinyurl.com/2e3tnq
          Flickr: flickr.com/photos/vrdundee/
          Youtube: youtube.com/user/VRDundee Member of IVRPA.org (International VR
          Photography Association): ivrpa.org/user/1133
          Member of the Panorama Tools Meeting Group Europe : panotools-meeting.com/
          Contributor to the WorldWidePanorama.com Project ( University of
          California Berkeley) : tinyurl.com/233lkg
          Member of 360cities.net network VR portal
        • AYRTON - avi
          ... Milko that s exactly that Not a cylindrical projection The client told me that the photographer took the photos on his hands just turning aroound on a
          Message 4 of 9 , Nov 1, 2007
          • 0 Attachment
            On 11/1/07, Milko Amorth <panotools@...> wrote:
            > Hi Michel,
            >
            > > I am not sure that this is perfectly accurate as my assumptions may
            > > be wrong. But it looks good;-)
            >
            > Interesting forensic.
            > I would have guessed over 180° fov. To me it does not look like a
            > cylindrical projection at all. It looks more like an array of horizontal
            > rectlinear tiles fudged together, rather unsuccessfully in the ceiling
            > area. Again, pure speculation, but fun.

            Milko that's exactly that
            Not a cylindrical projection
            The client told me that the photographer took the photos on his hands
            just turning aroound on a normal tripod and at the client's agency
            some young designer put all the shots together ...

            But he can not remember if the photographer turn 180º or more, or even less ...

            AYRTON


            >
            > Cheers, Milko
            >
            >
            >
            > --
            > Milko K. Amorth
            > 360° Immersive Imaging Productions
            >
            > VRCanada.ca
            > Vancouver.360cities.net
            > Ph:604.561.5101
            > fx: 604.909.5125
            >
            > Skype: VRdundee
            > Adobe Photographers Directory: tinyurl.com/2e3tnq
            > Flickr: flickr.com/photos/vrdundee/
            > Youtube: youtube.com/user/VRDundee Member of IVRPA.org (International VR
            > Photography Association): ivrpa.org/user/1133
            > Member of the Panorama Tools Meeting Group Europe : panotools-meeting.com/
            > Contributor to the WorldWidePanorama.com Project ( University of
            > California Berkeley) : tinyurl.com/233lkg
            > Member of 360cities.net network VR portal
            >
            >
            > --
            >
            > Yahoo! Groups Links
            >
            >
            >
            >


            --
            AYRTON 21-9982.6313 www.ayrton.com
            Ladeira de Nossa Senhora, 214 / sl. 101 www.vrfolio.com
            Outeiro da Glória - RJ - 22211-100 - Brasil www.vr-images.com
            Panoramas do Rio de Janeiro www.rio360.com.br
          • Alan Ball
            ... That guess looks about right. The earlier guess of 130 degrees seemed to be too narrow. I came to my own back-of-the-envelope guess from the square
            Message 5 of 9 , Nov 2, 2007
            • 0 Attachment
              --- In PanoToolsNG@yahoogroups.com, michel thoby <thobymichel@...> wrote:
              >
              > Hi Ayrton,
              >
              > I think the answer is about 167 degrees.
              >
              > As you requested it, I have tried to develop a technique to address
              > this unusual specific problem.

              That guess looks about right. The earlier guess of 130 degrees seemed
              to be too narrow. I came to my own "back-of-the-envelope" guess from
              the square ceiling and floor tiles. Barring a major interruption in
              the square grid, the angle is definitely fairly close to 180 degrees
              but isn't quite there. It is then a matter of estimating how many
              squares' depth correspond to one squares' width at the edges, then
              calculating the arctangent of the width-to-depth ratio to get the
              angle at each edge compared to the depth. It is clear that the view
              isn't quite 180 degrees at each side, so you subtract the sum of these
              calculated angles from 180 degrees to get the horizontal angle of view.

              Alan Ball
              scoundrel1728
            • Erik Krause
              ... That reminds me of my experiments how to achieve the orientation plate (arc) projection. Rik Littlefield helped me then to undestand the maths. Perhaps you
              Message 6 of 9 , Nov 2, 2007
              • 0 Attachment
                On Thursday, November 01, 2007 at 18:41, michel thoby wrote:

                > I am not sure that this is perfectly accurate as my assumptions may
                > be wrong. But it looks good;-)

                That reminds me of my experiments how to achieve the orientation
                plate (arc) projection. Rik Littlefield helped me then to undestand
                the maths.

                Perhaps you could use it to verify your assumption:
                http://wiki.panotools.org/Unusual_remappings

                best regards
                --
                http://www.erik-krause.de
              Your message has been successfully submitted and would be delivered to recipients shortly.