scaber wrote:I notice that Autopano shows your lens as a 24mm fisheye but the shots don't look particularly like a fisheye (mine show a little vignetting in the corners) so maybe just change the settings so it shows as a 24mm standard lens and see how that goes.
mediavets wrote:
He says it was a Sigma 15mm fisheye, on a Canon 650D with 1.6x cop factor that's equiv. to 24mm on a fullframe sensor.
There are any number of peculiarities about this image set.
1. Why would one choose a 15mm fisheye on a cropped sensor body?
scaber wrote:mediavets wrote:
He says it was a Sigma 15mm fisheye, on a Canon 650D with 1.6x cop factor that's equiv. to 24mm on a fullframe sensor.
There are any number of peculiarities about this image set.
1. Why would one choose a 15mm fisheye on a cropped sensor body?
Yes, it won't be acting like a fisheye lens on that sensor. Even when I used a 10mm lens I didn't select fisheye. Maybe I should go back and find those old images and select the fisheye option to see what results I get.
scaber wrote:mediavets wrote:
He says it was a Sigma 15mm fisheye, on a Canon 650D with 1.6x cop factor that's equiv. to 24mm on a fullframe sensor.
There are any number of peculiarities about this image set.
1. Why would one choose a 15mm fisheye on a cropped sensor body?
Yes, it won't be acting like a fisheye lens on that sensor. Even when I used a 10mm lens I didn't select fisheye. Maybe I should go back and find those old images and select the fisheye option to see what results I get.
so maybe just change the settings so it shows as a 24mm standard lens and see how that goes.
1. Why would one choose a 15mm fisheye on a cropped sensor body?
2. The images are shot landscape - this suggests he may not be using a proper pano head - in which case there's probably a lot of parallax..
3. It appears to be a single row in which case there may be too many images and too much overlap; 25-30% is enough, more is not better.
4. Apparently shot in Aperture priority with an aperture of f2.8 - that's all wrong.
The result is consistently perfect in full automatic mode with the newest beta version APG 3.5.2b2.
Yet, I got bad result looking like the OP's with earlier or venerable versions of Kolor's programs (APG 3.0.8 and... APP 2.6.8) even when the correct type of lens (i.e. fisheye) had to be corrected by hand before launching the detection!
user235123 wrote:2. The images are shot landscape - this suggests he may not be using a proper pano head - in which case there's probably a lot of parallax..
Yes, I don't have a pano head it was shot with a regular tripod. Nevertheless there shouldn't be problems like these.
Users browsing this forum: No registered users and 3 guests