Welcome Matrice Pilots!
Join our free DJI Matrice community today!
Sign up

X5 Barrel Distortion? Sample Picture Attached

Joined
Sep 15, 2015
Messages
20
Reaction score
6
Age
45
I've been shooting all weekend with the X5 and the stock DJI 15mm lens and I am pretty surprised to see barrel distortion in the footage. With such a narrow FOV I would not have expected that. It can be easily corrected in post, but it is very annoying to have to deal with. Attached is a sample picture showing the distortion.

And just like any fisheye lens, the more you angle the camera down, the more pronounced the barrel distortion is.

Just to make sure, I went and downloaded DJI's RAW sample footage, most of it was mountains or did not show the horizon, but in one of the RAW images it was definitely obvious where the ocean met the sky.

I checked the DJI site and it is interesting that they do not list the edge distortion value for their lens.
 

Attachments

  • BarrelDistortionExample.jpg
    BarrelDistortionExample.jpg
    508.6 KB · Views: 236
Last edited:
mmm...so I wonder if the Pano/Leica 15mm does the same thing?
When I got the Pro I thought there would never be a reason for me to not just use the stock lens, but I'm starting to wonder too if there isn't better glass for the X5 than the prime it ships with.

It would be very helpful if they posted ALL of their lens parameters including the ED% value so that we could better compare all of the specs.
 
When I got the Pro I thought there would never be a reason for me to not just use the stock lens, but I'm starting to wonder too if there isn't better glass for the X5 than the prime it ships with.

It would be very helpful if they posted ALL of their lens parameters including the ED% value so that we could better compare all of the specs.
2.4%
 
I've been shooting all weekend with the X5 and the stock DJI 15mm lens and I am pretty surprised to see barrel distortion in the footage. With such a narrow FOV I would not have expected that. It can be easily corrected in post, but it is very annoying to have to deal with. Attached is a sample picture showing the distortion.

And just like any fisheye lens, the more you angle the camera down, the more pronounced the barrel distortion is.

Just to make sure, I went and downloaded DJI's RAW sample footage, most of it was mountains or did not show the horizon, but in one of the RAW images it was definitely obvious where the ocean met the sky.

I checked the DJI site and it is interesting that they do not list the edge distortion value for their lens.

It's the rare (and expensive) wide angle lens that doesn't have some amount of barrel distortion. The M.Zuiko 12 mm has it too. As you said, it is correctible in post, although a pain in the a**. Photoshop CC now has the ability to do lens correction before sending it to Premiere.
 
It's the rare (and expensive) wide angle lens that doesn't have some amount of barrel distortion. The M.Zuiko 12 mm has it too. As you said, it is correctible in post, although a pain in the a**. Photoshop CC now has the ability to do lens correction before sending it to Premiere.

The part that is surprising is that it has more distortion than the P3 even though the P3 has a wider FOV. I'm thinking because the sensor is bigger it uses the full lens and shows its distortion more.
 
Normally MFT camera (eg. Olympus OM-D) would be expected to carry out lens correction 'in camera', so it is disappointing the DJI has not done this. I have raised this issue with Adobe and they replied that this was their expectation. Should this not be the case, profiles can be created for all approved lens for the X5 for Photoshop as is the case with Nikon and Canon products.
 
I've been shooting all weekend with the X5 and the stock DJI 15mm lens and I am pretty surprised to see barrel distortion in the footage. With such a narrow FOV I would not have expected that. It can be easily corrected in post, but it is very annoying to have to deal with. Attached is a sample picture showing the distortion.

And just like any fisheye lens, the more you angle the camera down, the more pronounced the barrel distortion is.

Just to make sure, I went and downloaded DJI's RAW sample footage, most of it was mountains or did not show the horizon, but in one of the RAW images it was definitely obvious where the ocean met the sky.

I checked the DJI site and it is interesting that they do not list the edge distortion value for their lens.
Thank you for sharing... interesting!
 
I've been shooting all weekend with the X5 and the stock DJI 15mm lens and I am pretty surprised to see barrel distortion in the footage. With such a narrow FOV I would not have expected that. It can be easily corrected in post, but it is very annoying to have to deal with. Attached is a sample picture showing the distortion.

And just like any fisheye lens, the more you angle the camera down, the more pronounced the barrel distortion is.

Just to make sure, I went and downloaded DJI's RAW sample footage, most of it was mountains or did not show the horizon, but in one of the RAW images it was definitely obvious where the ocean met the sky.

I checked the DJI site and it is interesting that they do not list the edge distortion value for their lens.

Is the barrel distortion the curvature of the earth on the horizon?
 
Is the barrel distortion the curvature of the earth on the horizon?

Barrel distortion is the fact that even a rectilinear (not fisheye) lens has a certain distortion tthat has the effect of slightly bulging out the center of the image. See diagram.

It has nothing to do with the Earth's curvature, it just looks like that when you tilt the lens down.
Barrel-Distortion.png
 
  • Like
Reactions: The Editor
All the MFT bodies have build in distortion fix on the fly, so it was surprising to see DJI didn't do that on X5, disappointed
 
I'm getting very little distortion with mine in PS-CR 9.2

There is an unusual (in my experience) message saying there is a built in lens profile that has been applied DJI F550 20.5mm

On Saturday I did some comparison photos with the X5/15mm and Nikon D700/24-70 f2.8 set at approx 30mm and I was pretty please with the comparison, especially considering the cost of the two systems was very similar eg £3500. These were shot at optimised conditions on mixed subjects and processed with a similar range of settings.

Every image needs to be processed for the intended output/use so there is no one size fits all in regard to that but these were processed for comparison purposes and should not be taken as a statement of best quality of either set up.

I'm writing this on an iPad with no access to the Dropbox I used but for anyone interested in looking at the raw files they can be found in this post in rcg

http://www.rcgroups.com/forums/showpost.php?p=32961592&postcount=1202
 

Members online

Forum statistics

Threads
2,727
Messages
25,387
Members
5,600
Latest member
RomanChrz