Blurred result when stitching one image after another

[cxy-ccc]

When i try to stitch 20 images, an error occurred. The code only stitched 16 images.
Traceback (most recent call last):
File "/home/zx/tiegui/stitching-main/setup.py", line 53, in
panorama2 = stitcher2.stitch([imgA,imgB])
File "/home/zx/tiegui/stitching-main/stitching/stitcher.py", line 92, in stitch
imgs, features, matches = self.subset(imgs, features, matches)
File "/home/zx/tiegui/stitching-main/stitching/stitcher.py", line 138, in subset
matches,
File "/home/zx/tiegui/stitching-main/stitching/subsetter.py", line 28, in subset
indices = self.get_indices_to_keep(features, matches)
File "/home/zx/tiegui/stitching-main/stitching/subsetter.py", line 53, in get_indices_to_keep
raise StitchingError("No match exceeds the " "given confidence threshold.")
stitching.stitching_error.StitchingError: No match exceeds the given confidence threshold.
But the image has enough overlapping region.
The result of 16 images and the first half is vague.

When stitch this image, the error occurred.

[lukasalexanderweber]

Can you share the images?

[lukasalexanderweber]

panorama2 = stitcher2.stitch([imgA,imgB]) dont you stitch just two Images?

[cxy-ccc]

• [ ]

I stitch one piece after another. The images are arranged in order.
I have no idea why the first half is vague.

[lukasalexanderweber]

Can't you just collect all Images in a List and than Stitch?

[cxy-ccc]

Can't you just collect all Images in a List and than Stitch?

This may cost more time. Now the error was solved, but i don't know why the first half is so vague.

[Greatuso]

Hi,I met the same situation,have you found the cause of this problem？

[lukasalexanderweber]

Try Just for the image quality to Stitch all Images together, so we can see If its vague as well

[cxy-ccc]

Try Just for the image quality to Stitch all Images together, so we can see If its vague as well

The quality is high, 1280*1280. And I didn't compress the size of image.

[lukasalexanderweber]

Can you share the images?

[lukasalexanderweber]

You can reduce the image size for testing with the final_megapix parameter. So you don't have to wait to long for a result

[KeineAhnung1337]

Hi, I have the same issue.
If I put all into a list, it takes ages to calculate and gets stuck at 4th step (03_matches_graph is made).
If i do it in intervalls of two recursive (e.g. 80 pictures into 40 into 20 into 10 into 5 into 2+1). If i put the list of 5 into the function, it crashes as well caused by a camera issue.
It is very blurry since I let it stitch for 5 times.
Is there any fix/parameter I can change? What are the detector options? I did not find anything in the docu.

[lukasalexanderweber]

Can you share sample Images?