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

Dizzying Panoramas Of Stars In Motion

Expand Messages
  • Peter A. Schaible
    Can anyone explain how these were made? http://www.huffingtonpost.com/2014/06/20/planetary-panoramas-vincent-brady_n_5515505.html I am experienced both at
    Message 1 of 13 , Jun 23, 2014
      Can anyone explain how these were made?

      http://www.huffingtonpost.com/2014/06/20/planetary-panoramas-vincent-brady_n_5515505.html

      I am experienced both at "night" photography and at making "little
      planets," but these are something else! Do you think there is a way to
      make them that doesn't require four cameras and four fisheye lenses?

      Your thoughts?

      --
      -- Peter

      Peter A. Schaible



      www.sundance360.com
    • Trausti Hraunfjord
      Vincent details his workflow here: http://www.vincentbrady.com/planetarypanoramas You can also communicate with him directly here:
      Message 2 of 13 , Jun 23, 2014
        Vincent details his workflow here: http://www.vincentbrady.com/planetarypanoramas

        You can also communicate with him directly here: https://www.facebook.com/groups/panoramicphotographers/

        Trausti


        On Mon, Jun 23, 2014 at 4:47 PM, 'Peter A. Schaible' peter@... [PanoToolsNG] <PanoToolsNG@yahoogroups.com> wrote:
         

        Can anyone explain how these were made?

        http://www.huffingtonpost.com/2014/06/20/planetary-panoramas-vincent-brady_n_5515505.html

        I am experienced both at "night" photography and at making "little
        planets," but these are something else! Do you think there is a way to
        make them that doesn't require four cameras and four fisheye lenses?

        Your thoughts?

        --
        -- Peter

        Peter A. Schaible

        www.sundance360.com


      • paul womack
        ... What s going wrong? - he s stacking his hundreds of frames into only four star-trail fish eye shots, and then stitching the pano from those four frames.
        Message 3 of 13 , Jun 24, 2014
          Trausti Hraunfjord trausti.hraunfjord@... [PanoToolsNG] wrote:
          >
          >
          > Vincent details his workflow here: http://www.vincentbrady.com/planetarypanoramas
          >
          > You can also communicate with him directly here: https://www.facebook.com/groups/panoramicphotographers/

          What's going wrong? - he's stacking his hundreds of frames into only four star-trail
          fish eye shots, and then stitching the pano from those four frames.

          He's manually placing (fair enough) 30 to 40 control points, and yet says
          "this usually takes me several hours."

          At 30-40 CPs per seam, that's 160 CPs in (say) 5 hours, 1.8 minutes per CP.

          That's not my experience of manual CPs at all.

          BugBear
        • Trausti Hraunfjord
          Seriously: Ask him, not me. The results of his works speaks louder than anything anyone in the panoramic world has ever managed to utter ... EVER ... when it
          Message 4 of 13 , Jun 24, 2014
            Seriously: Ask him, not me.  The results of his works speaks louder than anything anyone in the panoramic world has ever managed to utter ... EVER ... when it comes to startrail panos. Personally I don't care if he used 1 second or 3 weeks to place a single control point... what he has done, is exceptionally nice.


            On Tue, Jun 24, 2014 at 3:13 AM, paul womack pwomack@... [PanoToolsNG] <PanoToolsNG@yahoogroups.com> wrote:
             

            Trausti Hraunfjord trausti.hraunfjord@... [PanoToolsNG] wrote:
            >
            >
            > Vincent details his workflow here: http://www.vincentbrady.com/planetarypanoramas
            >
            > You can also communicate with him directly here: https://www.facebook.com/groups/panoramicphotographers/

            What's going wrong? - he's stacking his hundreds of frames into only four star-trail
            fish eye shots, and then stitching the pano from those four frames.

            He's manually placing (fair enough) 30 to 40 control points, and yet says
            "this usually takes me several hours."

            At 30-40 CPs per seam, that's 160 CPs in (say) 5 hours, 1.8 minutes per CP.

            That's not my experience of manual CPs at all.

            BugBear


          • paul womack
            ... Have I misunderstood the purpose of a discussion list? BugBear
            Message 5 of 13 , Jun 24, 2014
              Trausti Hraunfjord trausti.hraunfjord@... [PanoToolsNG] wrote:
              >
              >
              > Seriously: Ask him, not me. The results of his works speaks louder than anything anyone in the panoramic world has ever managed to utter ... EVER ... when it comes to startrail panos. Personally I don't care if he used 1 second or 3 weeks to place a single control point... what he has done, is exceptionally nice.

              Have I misunderstood the purpose of a discussion list?

              BugBear
            • paul womack
              ... Ah; I d only looked at the still shots on the page, not the video. It looks like he s stitching (and CP ing) each frame of the video from 4 star-stax d
              Message 6 of 13 , Jun 24, 2014
                paul womack pwomack@... [PanoToolsNG] wrote:
                > Trausti Hraunfjord trausti.hraunfjord@... [PanoToolsNG] wrote:
                >>
                >>
                >> Vincent details his workflow here: http://www.vincentbrady.com/planetarypanoramas
                >>
                >> You can also communicate with him directly here: https://www.facebook.com/groups/panoramicphotographers/
                >
                > What's going wrong? - he's stacking his hundreds of frames into only four star-trail
                > fish eye shots, and then stitching the pano from those four frames.
                >
                > He's manually placing (fair enough) 30 to 40 control points, and yet says
                > "this usually takes me several hours."
                >
                > At 30-40 CPs per seam, that's 160 CPs in (say) 5 hours, 1.8 minutes per CP.
                >
                > That's not my experience of manual CPs at all.

                Ah; I'd only looked at the still shots on the page, not the video. It looks like he's
                stitching (and CP'ing) each frame of the video from 4 star-stax'd images separately.

                I don't understand why.

                Surely using a template project, based on a sample 4 image set would provide a transform
                set for all the frames from a given camera position, which would then be scriptable?

                BugBear
              • Trausti Hraunfjord
                Personally I have never attempted anything even remotely related to what Vincent has done... but I would assume that a template in PTGui should really be the
                Message 7 of 13 , Jun 24, 2014
                  Personally I have never attempted anything even remotely related to what Vincent has done... but I would assume that a template in PTGui should really be the easiest and best choice... but then again... a lot of bright stars in the sky are moving between frames, so maybe that is the reason why he can't use a template  that easily?

                  I am not encouraging silencing of the original question... I rather want to see all questions asked and answered. I am simply not capable of providing the needed answers ... therefore I suggested taking direct contact with the creator of the content in question.


                  On Tue, Jun 24, 2014 at 5:20 AM, paul womack pwomack@... [PanoToolsNG] <PanoToolsNG@yahoogroups.com> wrote:
                   

                  paul womack pwomack@... [PanoToolsNG] wrote:
                  > Trausti Hraunfjord trausti.hraunfjord@... [PanoToolsNG] wrote:
                  >>
                  >>
                  >> Vincent details his workflow here: http://www.vincentbrady.com/planetarypanoramas
                  >>
                  >> You can also communicate with him directly here: https://www.facebook.com/groups/panoramicphotographers/
                  >
                  > What's going wrong? - he's stacking his hundreds of frames into only four star-trail
                  > fish eye shots, and then stitching the pano from those four frames.
                  >
                  > He's manually placing (fair enough) 30 to 40 control points, and yet says
                  > "this usually takes me several hours."
                  >
                  > At 30-40 CPs per seam, that's 160 CPs in (say) 5 hours, 1.8 minutes per CP.
                  >
                  > That's not my experience of manual CPs at all.

                  Ah; I'd only looked at the still shots on the page, not the video. It looks like he's
                  stitching (and CP'ing) each frame of the video from 4 star-stax'd images separately.

                  I don't understand why.

                  Surely using a template project, based on a sample 4 image set would provide a transform
                  set for all the frames from a given camera position, which would then be scriptable?

                  BugBear


                • mrjimbo2
                  The science of how to is always interesting.. For me I would likely not take something like this on.. I did however really very much enjoy viewing his work.. I
                  Message 8 of 13 , Jun 24, 2014
                    
                    The science of how to is always interesting.. For me I would likely not take something like this on.. I did however really very much enjoy viewing his work.. I feel it is stunning , innovative and somewhat unique.. The concept of stacking 360's followed by turning many into a video with the subject matter he has chosen.. Incredibly dramatic and certainly most enjoyable to view.. He's a visual artist for sure.. just sayin.........
                     
                    jimbo
                     
                    ----- Original Message -----
                    Sent: Tuesday, June 24, 2014 4:20 AM
                    Subject: Re: [PanoToolsNG] Dizzying Panoramas Of Stars In Motion

                     

                    paul womack pwomack@... [PanoToolsNG] wrote:
                    > Trausti Hraunfjord trausti.hraunfjord@... [PanoToolsNG] wrote:
                    >>
                    >>
                    >> Vincent details his workflow here: http://www.vincentbrady.com/planetarypanoramas
                    >>
                    >> You can also communicate with him directly here: https://www.facebook.com/groups/panoramicphotographers/
                    >
                    > What's going wrong? - he's stacking his hundreds of frames into only four star-trail
                    > fish eye shots, and then stitching the pano from those four frames.
                    >
                    > He's manually placing (fair enough) 30 to 40 control points, and yet says
                    > "this usually takes me several hours."
                    >
                    > At 30-40 CPs per seam, that's 160 CPs in (say) 5 hours, 1.8 minutes per CP.
                    >
                    > That's not my experience of manual CPs at all.

                    Ah; I'd only looked at the still shots on the page, not the video. It looks like he's
                    stitching (and CP'ing) each frame of the video from 4 star-stax'd images separately.

                    I don't understand why.

                    Surely using a template project, based on a sample 4 image set would provide a transform
                    set for all the frames from a given camera position, which would then be scriptable?

                    BugBear

                    No virus found in this message.
                    Checked by AVG - www.avg.com
                    Version: 2014.0.4592 / Virus Database: 3972/7731 - Release Date: 06/23/14

                  • Erik Krause
                    ... Of course not, but you ask a question only he can answer... -- Erik Krause http://www.erik-krause.de
                    Message 9 of 13 , Jun 24, 2014
                      Am 24.06.2014 11:39, schrieb paul womack:
                      > Have I misunderstood the purpose of a discussion list?

                      Of course not, but you ask a question only he can answer...

                      --
                      Erik Krause
                      http://www.erik-krause.de
                    • Erik Krause
                      ... If you use a full circular fisheye pointing straight up you can shoot the whole sky at once without the need to stitch the star trails itself. This way the
                      Message 10 of 13 , Jun 24, 2014
                        Am 23.06.2014 23:47, schrieb Peter A. Schaible:
                        > Do you think there is a way to
                        > make them that doesn't require four cameras and four fisheye lenses?

                        If you use a full circular fisheye pointing straight up you can shoot
                        the whole sky at once without the need to stitch the star trails itself.
                        This way the resolution would be limited, but you'll omit the problem of
                        star trails being interrupted along the seams.

                        Another way would be to to stitch the trails separately such that the
                        end of a trail would align to the start of the same star's trail in the
                        next image. This way time would be the limiting factor, if you want to
                        shoot all images in one night and you would be limited to plain streaks
                        (no saucer or comet effect, like in Vincent's images).

                        And last you could create artificial star trails. If you shoot the night
                        sky with relatively short exposure such that the stars are points and
                        orient the resulting panorama such that the north star (Polaris) is in
                        the zenith you can use photoshop motion blur filter to create horizontal
                        streaks in the equirect image. Once reprojected to normal orientation
                        you get trails rotating around the north star. This is more or less the
                        opposite of reducing star trails to star images like shown in
                        http://www.panotools.org/dersch/startrail/trail.html
                        Of course you'd need a night sky panorama which covers both the sky at
                        the beginning of the "trails" and at their end in order to have full
                        "trails" for all stars (even those who rise after your first shot). But
                        it might be enough to shoot one panorama soon after dusk and one just
                        before dawn and stitch them such that you get most of the sky.

                        For the last two methods the landscape needs to be overlayed with a mask
                        of course.

                        --
                        Erik Krause
                        http://www.erik-krause.de
                      • Keith Martin
                        ... Actually, what Vincent Brady has created is *very* impressive, but that doesn t mean the technical methods couldn t be improved. First of all, a stitcher
                        Message 11 of 13 , Jul 29, 2014
                          > schrieb paul womack:
                          >> Have I misunderstood the purpose of a discussion list?


                          Erik Krause wrote:
                          > Of course not, but you ask a question only he can answer...


                          Actually, what Vincent Brady has created is *very* impressive, but that
                          doesn't mean the technical methods couldn't be improved.

                          First of all, a stitcher template would be eminently sensible, and it
                          would mean stitching could be done without further control points. Sure,
                          things are moving in the sky, but that's irrelevant; the shots are all
                          from the same positions so each set can be stitched without having to
                          fuss with CPs once the correct values have been set up once.

                          Second, judging by the shot at the start that shows a large metal '+'
                          with the cameras mounted at the end of each arm, the cameras are much
                          further offset than they need to be. Moving them as close together as
                          possible would make the output more reliable.

                          Creatively? Lovely stuff.

                          k
                        • Erik Krause
                          ... You could even use a one shot solution. A circular fisheye pointed straight up covers the whole sky, which would be perfect for long exposures. And since
                          Message 12 of 13 , Jul 29, 2014
                            Am 29.07.2014 18:12, schrieb Keith Martin:
                            > Second, judging by the shot at the start that shows a large metal '+'
                            > with the cameras mounted at the end of each arm, the cameras are much
                            > further offset than they need to be. Moving them as close together as
                            > possible would make the output more reliable.

                            You could even use a one shot solution. A circular fisheye pointed
                            straight up covers the whole sky, which would be perfect for long
                            exposures. And since those comet-like star trails are more or less
                            artificial anyway (normal star trails don't fade), one could think about
                            completely artificial trails.

                            As Helmut Dersch showed years ago
                            <http://www.panotools.org/dersch/startrail/trail.html> star trails are
                            straight and equally long in an equirectangular projection with the
                            zenith at polaris. So it should be easy to create "star trails" by a
                            simple motion blur of a sky panorama. (In fact it's not so easy, since
                            the trails darken due to blur. You need to duplicate layer with mode
                            lighten, shift 1px and flatten). This would reduce shooting to start and
                            end of the night in order to also cover stars that rise during the night.

                            Furthermore stars are colored. This is seldom visible, since stars show
                            as points which are overexposed very quickly while other stars are not
                            even visible. If you defocus slightly, stars are more like disks, which
                            don't overexpose that fast. Could be light pollution prohibits this - I
                            discovered it on altiplano in Peru 4000m above sea level and hundreds of
                            miles away from any city, one of the darkest regions on earth.

                            --
                            Erik Krause
                            http://www.erik-krause.de
                          • Jim Watters
                            ... http://www.tawbaware.com/startracer.htm -- Jim Watters http://photocreations.ca
                            Message 13 of 13 , Jul 29, 2014
                              On 2014-07-29 2:26 PM, Erik Krause erik.krause@... [PanoToolsNG] wrote:
                              > You could even use a one shot solution. A circular fisheye pointed straight up
                              > covers the whole sky, which would be perfect for long exposures. And since
                              > those comet-like star trails are more or less artificial anyway (normal star
                              > trails don't fade), one could think about completely artificial trails.

                              http://www.tawbaware.com/startracer.htm


                              --
                              Jim Watters
                              http://photocreations.ca
                            Your message has been successfully submitted and would be delivered to recipients shortly.