Seeking Expert for High-Resolution Aerial Panorama Stitching (189 Images)

97 views
Skip to first unread message

Kevin Henderson

unread,
Apr 30, 2026, 3:02:30 PMApr 30
to PTGui Support
  I am seeking an experienced image editor or panoramic compositing specialist to create a large-format, visually seamless aerial coastline composite from approximately 189 high-resolution images captured from a manned aircraft. The images were taken in a single left-to-right flight line along the Gulf Shores to Perdido Key area and include slight oblique angles with some horizon visible. The final deliverable must be visually accurate and cohesive, with roads, coastlines, islands, and bodies of water appearing continuous, complete, and naturally aligned across the entire image. There should be no visible stitching seams, breaks, or mismatches in features such as roads or shorelines, and buildings should appear natural without excessive leaning or distortion. While survey-grade accuracy is not required, the image must maintain strong visual integrity suitable for a large wall print up to 8 feet wide. The editor should be comfortable working with large datasets, breaking the project into sections if needed, and using advanced tools such as Photoshop, PTGui, or similar to manually refine alignment and blending. The final output must be high-resolution, color-consistent, and fully print-ready.  This is a work for hire project. 

Budget: $1,000, payable upon successful completion of a visually accurate, print-ready composite. The final image must meet all stated requirements for seamless blending, alignment, and overall visual integrity.  

John Houghton

unread,
May 1, 2026, 9:49:08 AMMay 1
to PTGui Support
Kevin, Based on a few similar, if smaller, projects seen over the years, I fear it may not be feasible to produce a stitch that fully meets your exacting requirements using PTGui.  However, I like a challenge and if you care to make available a sample subset of, say, six overlapping images, I would be happy to attempt a trial stitch to assess their stitchability.  Half size jpegs would be good enough.

John

Jesús Navas

unread,
May 1, 2026, 1:23:32 PMMay 1
to pt...@googlegroups.com
Hi Kevin, 

I could it give it a try too, if you wish.

I agree with John. Depending on the way the images were taken, it may be not possible to achieve your expected outcomes because of the geometry of the image even if the route of the plane and the coast line had been perfectly straight and parallel due to the oblique angle of the shots and the different distances to the sensor even in the very single shot. If the plane had flown at the same height and the camera had been parallel to the surface, then it could have a chance. But it is a very interesting project indeed. 

If it is a creative project, it may be achieve by dividing the panorama into several coherent pieces and showing them very close to give the impression of continuity. Maybe even in two rows, not only one. 

Could you share in a map (a Google Earth file better) the roughly route followed by the plane and the photographied coast, and 6 images from the begining and 6 images from the end of the panorama? 

Is there any deadline to finish the panorama? 

Jesús 

--
You received this message because you are subscribed to the Google Groups "PTGui Support" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ptgui+un...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/ptgui/103722b9-42f4-43a6-953c-d88b52ca5e60n%40googlegroups.com.

HDouris

unread,
May 1, 2026, 2:38:05 PMMay 1
to pt...@googlegroups.com
Kevin,

What is your definition for Hi-res ?
Try to use the « rotative camera » tech, used in the old times. Actually, this is doable using a 8k camera in portait or diagonal orientation and extract and stich smaller as possible columns of pixels.
Sorry not able for other references.

Romuald

sc...@highton.com

unread,
May 1, 2026, 8:22:13 PMMay 1
to PTGui Support
Kevin,

Yours may not be the best approach to a project like this, but you also don't provide sufficient details about how you shot the source images.

1)  What camera & lens combo did you use.  We need to know what the resolution of each image is, and what the lens focal length and camera sensor size were.
     Longer focal lengths are usually best for this sort of mosaic stitching, but they present their own challenges when being used from a moving aircraft or vehicles.  Note that oblique shooting angles exacerbate parallax issues between shots because of widely varying subject distances.

2)  How much subject overlap do you have between adjacent shots?
     For most panoramic stitching, 30–50% overlap works best, but for mosaic stitching MUCH more is preferred.

3)  Did your aircraft course (track) follow the coastline, or was it flown in a straight line?  Both would be the best combination (straight flight track following a straight coast line), but not usually possible.

4) Did you use a gyro stabilizer on your camera while shooting, and did you shoot out an open door or window, while keeping the camera and lens inside out of the air stream?
     All this points to how sharp, focused, and color accurate your source images are.

5)  Are your exposures consistent (along with the lighting) throughout the 189 images?

6)  Can your client accept multiple panels of the coast line, rather than a single eight foot wide "perfectly stictched" image?

Your end goal of a long panoramic image of the coastline and its features without stitching artifacts and distortions in seam areas – AND of sufficient quality to be enlarged to 8-feet wide is not unreasonable.  However, your approach of shooting 189 images and hoping to stitch them together was probably not your best approach.  Parallax differences between shots are going to be your biggest nightmare, and you (or whomever tries to assemble these) may struggle in post production for many days (for only $1,000) and still wind up with only mediocre results.  Some might be tempted to use AI imaging tools to "fix" the errors, but this will result in inaccurate representation of the coast pretty much lacking any visual integrity.

You might have been better off using a single exposure of the entire coastline section shot with an ultra-high resolution camera, or using a scanning (slit) aerial camera / service, rather than making things as complicated as you have.  Your chosen approach is probably going to be more time consuming, more expensive, and yield lower quality results than you're hoping for.  Unfortunately, your desired oblique perspective is likely to make a scanning or slit scan camera approach unworkable, however, because these generally need to align the images/scan speeds based on consistent camera-to-subject distances.

When shooting something new, particularly for a client, it's best to figure out your procedures (planning, equipment, shooting techniques, production, and post production) ahead of time, rather than  just assuming that if you shoot lots of images, something can be put together from them.  A forum like this is a good place to get help when planning your shoot, rather than after the fact of having made one or more poor choices.

Good luck, and hopefully you'll have a chance to do a re-shoot.  Or you might get lucky with one of the post-production magicians who reside here.  :-)

Scott Highton
Author, Virtual Reality Photography

Kristof Meirlaen (Kri-Soft)

unread,
May 11, 2026, 5:14:50 AM (10 days ago) May 11
to PTGui Support
If this is still open, I want to give it a shot :)

Op zaterdag 2 mei 2026 om 02:22:13 UTC+2 schreef sc...@highton.com:

Kevin Henderson

unread,
May 14, 2026, 12:40:19 PM (7 days ago) May 14
to PTGui Support

Yes, the project is still open. The photos were captured using a Sony A7V in JPG format. I used a G Master 24–70mm lens at around 40mm for 189 images, and on a completly separate pass a 70–200mm lens at 70mm.

All images were taken handheld, so they are not perfectly consistent in heading or angle from shot to shot. I captured approximately 189 images over roughly 19 miles, so there is overlap throughout the sequence, although some areas have more overlap than others due to manually triggering the shutter. Exposure is very consistent across the set.

My goal is not to produce a survey-grade result, but rather a wide, continuous image that is visually accurate and natural-looking. The final output is intended to be a large-format print for an office wall, so visual continuity and realism are the priority.

Reply all
Reply to author
Forward
0 new messages