Best method you know for finding nodal / parallax point / entrance pupil ?

[Aristoc]

What is the best method you know for finding your nodal point (aka parallax point , entrance pupil)?

I just used this method on this persons web page. It's a piece of tape on the window and then you focus on something far off into the distance. I just tried it and it seemed to work. Checked the images in photoshop to see if they overlapped.


see #4 lower on his page:
http://www.johnhpanos.com/epcalib.htm

[Joe Behar]

You can do something very similar in the field as well.

Line up two vertical objects, such as telephone poles. One close to the camera, the second further away. as you pan the camera, see if the two poles stay aligned. if they do, you're on the nodal point.

[rljones]

The entrance pupil is where the diaphragm appears to be as you look into the front of the lens. By stopping down the lens, you can estimate this location pretty easily.

More accurately, I use a vertical laser (like you can get at Home Depot; can project horizontal and vertical beam) projected into the front of the lens while on the camera with the diaphragm stopped down. The camera needs to be level on a pano head with the laser on an adjacent table. If the fore/aft adjust of camera is at the exact entrance pupil, the beam on the diaphragm will appear to be stationary as you rotate the camera side to side. If behind or in front of the entrance pupil, the beam will appear to move with or against the direction of side to side camera rotation.

As an aside, I did this for a series of lenses for my M9 and found that the entrance pupil varies with lens focus. Since we are typically interested in infinity focus for most panos, this is not an issue (although it does mean you need to verify that lens is at infinity focus when you perform the above measurements). However, if you wish to stitch near objects (still life, etc), then you need to adjust the entrance pupil for that distance.

[OldRoy]

John Houghton's tutorials on pano techniques are excellent. Here's a link related to your question that may be useful.
http://www.johnhpanos.com/epcalib.htm
Roy

[Aristoc]

Once you have tested your camera and think you found the parallax point...how can you check , after combining final images, that you did a good job?

[Tim Gray]

First, parallax is mostly an issue for wide angles with foreground.  So to check, just create a pano of several shots then in PS create a canvass big enough and manually drop the images into the canvass.  A brick driveway in the foreground would be a good test.  If you are right with the offset the images should align pretty well with all lines continuous, not broken from one image to the next.   If you use one of the pano merge options in PS then if there's a problem, it will be clear from the way the lines break or ghost in the final image.

[Bart_van_der_Wolf]

Once you have tested your camera and think you found the parallax point...how can you check , after combining final images, that you did a good job?

You need to test it before you combine the images. Just look at foreground and background features and compare images that are rotated left and right, and/or up and down. Do the foreground and background features move relative to one another.

After stitching you can only test it by producing a layered result, layers still intact, but that result also includes residual  software (stiching) errors. So it is best to test before stitching, there you are testing for parallax only.

Cheers,
Bart

[elf]

The entrance pupil is where the diaphragm appears to be as you look into the front of the lens. By stopping down the lens, you can estimate this location pretty easily.

More accurately, I use a vertical laser (like you can get at Home Depot; can project horizontal and vertical beam) projected into the front of the lens while on the camera with the diaphragm stopped down. The camera needs to be level on a pano head with the laser on an adjacent table. If the fore/aft adjust of camera is at the exact entrance pupil, the beam on the diaphragm will appear to be stationary as you rotate the camera side to side. If behind or in front of the entrance pupil, the beam will appear to move with or against the direction of side to side camera rotation.

As an aside, I did this for a series of lenses for my M9 and found that the entrance pupil varies with lens focus. Since we are typically interested in infinity focus for most panos, this is not an issue (although it does mean you need to verify that lens is at infinity focus when you perform the above measurements). However, if you wish to stitch near objects (still life, etc), then you need to adjust the entrance pupil for that distance.

You can measuring the apparent position of the diaphragm fairly accurately using a macro lens with a narrow depth of field. Put the lens on a table top, mark where the front (or back) of the lens is, focus the macro lens on the camera on the diaphragm of the lens being measured, then replace the lens with a block a wood (or anything that is 90 degrees), move it until the front face is in focus, measure the distance.

I like your laser technique better than the ones that have you drawing lots of lines on a piece of paper.  I have a couple of lens (microscope objectives) that don't have a visible diaphragm so I can't use the macro lens technique on them. 

Once you have tested your camera and think you found the parallax point...how can you check , after combining final images, that you did a good job?

Take three shots of any subject (it's best to have 3 recognizable features in the foreground, middle , and background). Use the Load files into stack... feature of PhotoShop (don't align them at this point).  Manually align the images on the focus point. Toggle the visibility of the layers on and off.  If there is any movement visible, adjust the panohead and reshoot.

My 70-300mm lens' entrance pupil changes from near the front of the lens at 70mm to more than 18 inches behind the camera at the 300mm close focus distance.  I'm still trying to figure out how it's possible for the light to exit the lens in front of where it enters the lens.

[01af]

... your nodal point (aka parallax point , entrance pupil) ...

The nodal point is not "also known as" the entrance pupil or no-parallax point. The no-parallax point for panorama-shooting purposes is where the entrance pupil and the optical axis intersect. The nodal points (two of them!) are entirely different things.

[Aristoc]

What does that mean?

[elf]

It just means that you should use entrance pupil or no-parallax point when discussing panoramas. The front nodal point may or may not be at the same position as the entrance pupil and has a different meaning optically.  It's just a matter of retraining your tongue to say 'entrance pupil' when you're thinking 'nodal point'   Most people will understand what you mean, but it's best to use the correct terms.

[Aristoc]

Ill  have to remember why I said that in the first place. Reading all over.

TY