PanoTools:
Re: helicopter panorama - shooting method
Bernhard Vogl 2005-Jun-11 14:30:48
Hello Edward,
I've played around with the images you provided and i am sorry to tell
you that there is no chance to stitch a good panorama with these.
Basic considerations:
- This is an architectural shot. There is no room for "creative adaption"
- Most important image areas are the horizon and the nadir area. If you
you can't stitch these, you're lost.
- My idea of a matching workflow was, to stitch a good horizon-area down
to approx. -50 to -60 degrees. For the area below that i intended to fit
in a separate nadir shot.
So what went wrong?
The failures of the pilot have already been discussed at large. He did
fly a curve with inreasing radius. I doubt that he did this for security
reasons: the speed would have been still to low to switch to
autorotation if the engine fails...
What are the failures of the photographer? (I hope you don't take this
too personal)
- You shot many images (they are more than enough) but you didn't finish
the "overshooting-circle". There is a "big" gap between the 0 and 360
degree area, compared to the other images. This makes it difficult to
calculate lens parameters for a nicely fitting seam. Nethertheless - the
horizon area can be stitched flawlessly by completely ignoring the nadir
area.
- This makes it important to have a good nadir shot. The "nadir" images
you have, are not really nadir shots, as the skids of the helicopter are
in the way. Even worse, the most difficult part of this area is missing:
The skyscraper area.
So there isn't even a chance to assemble the nadir area with some of the
provided images.
_If_ you think about re-shooting with the helicopter, think about a way
to retrieve a good nadir shot. There seems to be enough room between the
heli and the skids to hold down a monopod in between (don't forget to
secure all parts of the equipment with a strap tied to the helicopter).
You may instruct the pilot to fly above your "nadir area" several times
to be able to select the best matching one afterwards.
Given the facts of the shootings so far, you may also think about a less
complicated way of shooting your panorama.
One idea that came into my mind: I was working for a small radio station
several years ago. For advertisment purposes we rent a small helium
balloon about 5 meters in diameter and we were allowed to rise the
balloon in the center of a large town.
The balloon is strong enough to carry the needed equipment: A remote
controlled rig (like the ones, the kite flyers use) hanging down on a
platform a few meters below and the camera. This solution is also cheap
enough to invest some time in testing and improving equipment and
workflow before. Afterwards, you may position yourself on top of one of
the skyscapers on a day with not too strong wind and aquire your images...
I can't guarantee that this would work for you - i also never tried it
before, but for the money invested for one hour heli-flying, you should
be able to buy the rig, some helium and rent the baloon for several
days... ;-)
Best regards
Bernhard
Edward Fink wrote:
> I have no intention of badmouthing the pilot, at least not until
>I can fly a helicopter better than he can. In fact, if the client
>decides to try again, I'd probably vote to use the same pilot again,
>just because I'm already familiar with him and his helicopter, and I
>know what to expect.
>
> But I would not have the pilot do the rotation for me next time.
>Clearly it isn't an easy task and right now I have no reason to think
>that this pilot or any other could definitely nail it on the next try.
>
> I finally got in touch with an old friend in Sweden that's a
>helicopter pilot, and he says it's the turning around that's causing the
>problems, especially since we weren't flying in zero wind. He says that
>in the time it took to shoot the series, we rotated from a headwind to a
>sidewind to a tailwind and another sidewind, which was almost certain to
>cause us to move. He said it would be far easier for any pilot to hover
>flying into the wind, without trying to rotate.
>
> It's also VERY difficult to see exactly what we were over. We're
>a half mile up. I was leaning way out and had a far better view than
>the pilot, and it was hard for me to tell if we were really over the
>building or the street, or the parking lot across the street, especially
>as we rotated. It seemed to vary depending on how far I was leaning out
>too. You almost need something like a bomb sight to tell for certain.
>
> My biggest complaint with the pilot is something that's not
>really his fault: he's not as enthusiastic about the project as I am.
>It seems to be just another flying job to him, rather than him being
>like a partner or team member actively working to find a solution. To
>me, the ideal pilot would be somebody who is excited about VR and wants
>to help make a killer aerial VR so he can get a copy for his website.
>
> Also, I should point out that I'm not in charge of the project,
>so it isn't up to me which of everyone's suggestions to implement. The
>client is an ex pilot (not helicopter), plus he's done cylindrical VR's
>before, so he's making a lot of the decisions about the flying and the
>photography. I pass the tips on, along with my opinion, but I'm not
>footing the bill so I don't get to call the shots. (Maybe I will
>though, because I can be very stubborn and independent about some
>things, and if push comes to shove, I just might pay for my own flight
>so that I'm free to try whatever I want, and to prove I can do it.)
>
> Next time, I only have faith in trying two options. One is to
>rent a camera, maybe a film camera, with a 1.0 crop factor so I can get
>a full 180 degree circle, and hang that from a tether or pole under the
>landing skid. The other is to put my D70 on a pole, either upside down,
>extended under the landing skids, or sideways several feet out past the
>landing skid (like Willy's monopod shots over the railing.) I
>downloaded Nikon Capture tonight and tomorrow I'll pick up a USB
>extension cable so I can fire the camera from my laptop for some pole
>tests.
>
> The pilot said he was "intrigued" by the idea of hanging a pole
>underneath, but on the other hand he seems nervous about it and hasn't
>really signed off on the idea yet. He did say there were no regulations
>against it as long as I didn't drop it on anyone's head. He said he
>likes to drop ping pong balls from the helicopter. :)
>
> I wouldn't mind trying some of the suggestions like making
>cloverleaf passes over the location, and/or having spotters on the
>ground, but because that leaves me too dependent on others, I think I'd
>only do it as an experiment after shooting with the pole.
>
>Ed
>
>-----Original Message-----
>From: #removed# [mailto:#removed#] On
>Behalf Of tflyfish2002
>Sent: Thursday, June 09, 2005 1:06 PM
>To: #removed#
>Subject: [PanoTools] Re: helicopter panorama - consider a different
>pilot
>
>With respect, I doubt if there is anything wrong with the pilot, but
>more with your technique and understanding of the problem. You have
>been offered several good solutions to your not being able to shoot a
>decent pano, have you tried any of them? You could do a dummy run from
>the top of a step ladder.
> Please don't diss someone who is a professional whilst you are just
>begining to understand the problem.
>
>Barry
>--- In #removed#, "Mark D. Fink" <#removed#> wrote:
>
>
>>This might be a good topic for a Wiki page - local ancillary
>>
>>
>services that
>
>
>>provide good (or bad) services. If anyone else happened to get an
>>
>>
>assignment
>
>
>>for aerial photography in your area, I'm sure the last thing they
>>
>>
>would want
>
>
>>is to hire this pilot and find out the hard way that his attitude
>>
>>
>gets in
>
>
>>the way of getting the job done.
>>
>>
>>
>>Mark
>>
>>www.northernlight.net <http://www.northernlight.net/>
>>
>>
>>
>>-----Original Message-----
>>From: #removed# [mailto:#removed#]
>>
>>
>On Behalf
>
>
>>Of Eric O'Brien
>>Sent: Thursday, June 09, 2005 12:46 AM
>>To: #removed#
>>Subject: [PanoTools] Re: helicopter panorama - consider a different
>>
>>
>pilot
>
>
>>
>>
>>On Jun 8, 2005, at 2:37 PM, Edward Fink wrote:
>>
>>
>>>>Also I wonder why you didn't take the time to direct the pilot over
>>>>
>>>>
>a
>
>
>>>vertical edge pointing to you as I suggested...
>>>
>>>The pilot doesn't appear to place much value in suggestions from
>>>
>>>
>mere
>
>
>>>mortals.
>>>
>>>
>>Umm...
>>
>>I think you may need a different pilot.
>>
>>You are not "a mere mortal" -- you are his CLIENT. How long would you
>>
>>
>
>
>
>>get away with treating your clients like that??
>>
>>
>>You are paying him money for a service. For that pay, he should be
>>VERY interested in accommodating your needs. If he feels that he
>>cannot accommodate your needs, he should be forthright about that.
>>Otherwise, he AT LEAST should be willing to work with you.
>>
>>
>>There must be more than one helicopter pilot in Minneapolis!
>>
>>eo
>>
>>[Non-text portions of this message have been removed]
>>
>>
>>
>>
>> _____
>>
>>Yahoo! Groups Links
>>
>>* To visit your group on the web, go to:
>>http://groups.yahoo.com/group/PanoTools/
>>
>>
>>* To unsubscribe from this group, send an email to:
>>#removed#
>><mailto:#removed#?subject=Unsubscribe>
>>
>>
>>* Your use of Yahoo! Groups is subject to the Yahoo! Terms of
>><http://docs.yahoo.com/info/terms/> Service.
>>
>>
>>
>>[Non-text portions of this message have been removed]
>>
>>
>
>
>
>
>
>Yahoo! Groups Links
>
>
>
>
>
>
>
>
>
>
>
>Yahoo! Groups Links
>
>
>
>
>
>
>
>
>
Yahoo! Groups Links
<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/PanoTools/
<*> To unsubscribe from this group, send an email to:
#removed#
<*> Your use of Yahoo! Groups is subject to:
http://docs.yahoo.com/info/terms/