Serene Life

by garik

submit your photo


Hall of Fame
View past winners from this year

Please participate in Meta
and help us grow.

Take the 2-minute tour ×
Photography Stack Exchange is a question and answer site for professional, enthusiast and amateur photographers. It's 100% free, no registration required.

I took a picture of a crescent arch with scaffolding (apparently for maintenance) but can't post it for lack of "10 reputation." However the effect is simple, in that from the vantage point, the row of scaffolding directly ahead appears as a single line, but those on either side show more of the "grid" of scaffolding. Same effect would be with any series of parallel rows of objects photographed from a point in line with one of the rows.

The question has to do with achieving an effect wherein the picture shows all rows head on? One could take a series of photos in line with each row (perhaps with multiple lenses) and photoshop them into one picture, which is perhaps similar to what Google does for streetview (have not found a detailed explanation of their technique). However that could be tedious for an individual with a camera and a specific photo subject, especially if there were lots of rows.

The question is whether there is a kind of rolling (as in moving) or incremental method to blend a very large number of very narrow head-on views of the rows and spaces between them? In other words adding slices of the visual field in as one moves rtl or ltr.

Hope this makes some sense - easier to visualize with the photo.

share|improve this question
1  
Welcome to Photo.SE! You now have >10 rep and should be able to post the picture. –  coneslayer Aug 4 '13 at 14:41
    
I think somebody has done what you want using Hugin: flickr.com/photos/36383814@N00/5830006193 –  coneslayer Aug 4 '13 at 14:45
    
If I understand what you want, Hugin can do it. It's not at all easy to use (requires some time to learn), but it's a very powerful and flexible panorama stitching program (it's flexibility is party to blame for its complexity...). Check out this tutorial first, then maybe this one –  Szabolcs Aug 4 '13 at 19:06
    
Talking heads did this also: See adayintheweek.files.wordpress.com/2008/03/talkingheads2.jpg –  Paul Cezanne Aug 4 '13 at 22:08

3 Answers 3

There are several ways to achieve this. You can manually take lots of pictures, and make sure all rows get one picture head on and then combine them in post.

Some cameras (not consumer or even prosumer) are pretty much made for this kind of work. This is 1D cameras, that record a single row of picture for each exposure. By moving the camera and taking pictures you can build a continuous panorama where everything is head on. The challenge is to move the camera the right amount and keep control over the angle. The easiest way to do this is to use a dolly rail, where you move the camera along fixed rails. If you match the speed along the rails with how often you take a new pixel row, you can get a nice looking photo.

This technique is also used from aircrafts, but then they rely on very accurate navigation and angular measurements and lots of post processing to get a nice result.

More information on strip photography on Wikipedia.

share|improve this answer
2  
Google the "buggy/wagon" with words dolly rail . –  Esa Paulasto Aug 4 '13 at 17:01
    
Thanks, @EsaPaulasto, I couldn't remember what they where called. –  Håkon K. Olafsen Aug 5 '13 at 5:02

You must stitch your images together using a rectilinear projection. Most, but not all, panorama programs let you choose this. For normal projection you describe is called cylindrical which is the default for most programs.

You can read and see examples about projections here and here. This nice thing is that this lets you work with images taken from a single vantage point and that makes it possible to resolve properly feature across images, assuming proper rotation around the nodal point of the lens.

The other technique you describe exits. Sony's Sweeping Panorama and Fuji's Motion Panorama work that way. Now if you do this while moving, you are in a situation where each slice may not match the others because of relative movement of objects in the scene. Smaller slices reduce the problem.

An effective way to get a low-resolution version of this would be to record a video while moving parallel to your subject and taking slices from each frame.

share|improve this answer

With Hugin you can do a lot of this things but is quite complex and sometimes frustrating. So usually I prefer others software. The simplest is ICE: you can correct lens distortion with Gimp LensFun before use ICE for better results. In this case you should take many photos with overlapping areas, and mantain the same parameters (iso,exposure,aperture...) from the first to the last. With ICE and Windows 7 you can make a video and than make a panorama from it. A little dolly is the best thing for this kind of works.

Check TrakEM2 also is a good software for scientific use.

share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.