Panorama Crop Apk

1 view

Skip to first unread message

Lauren Redder

unread,

Aug 3, 2024, 5:00:07 PM8/3/24

to pramibbrocpet

However, when I bring the two videos into either VideoStitch Studio or AutoPano Video I can get the two videos synchronized and calibrated, but I cannot figure out how to crop the projection to include the region of interest. (With VideoStitch Studio I even load the Hugin .pto file; I thought this should include the crop.) The videos they produce are always a small region of the video inside a sea of black:

(The final video dimensions will be very high resolution, on the order of 6k horizontal pixels, so I'd strongly prefer not to say "Just render it extra extra big and then use After Effects to crop out the center.")

Howdy...still learning the finer details of Affinity Photo...I'm using 1.7.2.471 on Windows...trying to post a panorama on Instagram using the multiple sequential photo technique. Danged if I can figure out the guide system for nicely creating equi-distance widths so I can crop the wide pano down into multiple images. I'm hoping there's a nice way to setup the guides on the pano, then have Affinity Photo crop down the segments into their own individual files. Lots of video tutorials for PS on how to do this, and of course I do my work in Affinity.

There may be some better way to do this but all I could think of is to manually create slices of the desired height, width, & position in the Export persona. This would be a lot simpler if there was a way to duplicate & reposition slices (like by alt-dragging on one with the Slice Tool) but I could not figure out how to do that.

I would do the same, although I don't think it's that necessary to position them using maths like that, you can simply drag them together and they'll snap nicely next to each other. Also once you've set the width on the first slice, you should be able to drag create the others and as you do so there will be a snap point at the same width as the previous slice, so you can very quickly just drag create the rest of them. It's fairly swift that way, you just need to keep an eye on the snapping.

How do you determine the width of the slices if not by using math of some sort? That was the only way I could figure out how to get all the slices to have equal widths, or for that matter to know how many slices were needed.

I could not get that to work. Not only was there no snap point for the same width for the next slice, I could not find any way to start dragging it out from the edge or corner of the previous slice -- all that would do is resize that slice.

You have to create the new slice to the right because as you say it doesn't like drawing it from the previous slice corner. You can then drag it back and snap it into place on the end of the previous slice.

I have just realised that I was in Designer, not Photo, and weirdly that behaviour is not consistent between the two (I wonder why?) so the width snap doesn't seem to work in Photo, but works fine in Designer. Apologies for that confusion!

I just realised you can draw it out from the previous slice if you deselect the previous slice first - it's just the transform handles that are getting in the way. It will snap to the position and the height in this way, leaving you to just set the width. In photo this time!

@R C-R Yes of course you'd need to use some maths to get the width if you wanted to split the entire picture into equal segments. I was more referring to the positioning not the width. Even without the width snap it's going to be easier to use the snapping to position the slices rather than typing the calc into the x position. So the workflow would be to work out the desired width by dividing the picture width by number of slices (or whatever method - maybe you have a desired slice width instead), then draw the first slice and set it's height and width the the required settings, then deselect the first slice, draw a new slice out from the top right corner of the first slice, until it snaps at the correct height, set the width, then deselect and repeat with the subsequent slices, setting the width each time. Not too bad, but much easier to do in Designer with it's better snapping to width feature.

I have just realised that I was in Designer, not Photo, and weirdly that behaviour is not consistent between the two (I wonder why?) so the width snap doesn't seem to work in Photo, but works fine in Designer.

But after watching the video @dutchshader posted carefully, I realized that it does work in the same way on Macs -- it is just necessary to deselect the current slice first to be able to drag out a new one from the corner of an existing one, & it does snap to the same width. (In the vid, I think that was done by clicking once somewhere not already in a slice.) For some stupid reason it never occurred to me to deselect the current slice first.

EDIT 2: I just realized that it isn't even necessary to to determine the appropriate slice width beforehand. Just snap the first slice to the full width & height of the image. Then, while it is still selected in the Transform panel add "/n" after the W value & press Return or Tab, where n is the desired number of slices. If this is done while any of the 3 left edge anchors are selected, it isn't even necessary to snap the slice to the left edge of the pano.

Sooo...thanks a bunch for the quick replies. It pointed me in the right direction...I realized that I knew zip-zero-nada about slices and export personas, and found the help video on that (link below), and was able to export 5 sequential snips of my pano to the right dimensions with the right cropping.

To share an example...remember we want to maximize IG's photo dimensions without causing a server side resizing...vertical is then 1350, never changes...max horizontal is 1080. So when cropping a wide image, first snip ends at 1080, second snip starts at 1081 and ends 1080 pixels later, third snip starts at 2162 and ends 1080 pixels later, etc.

I'm using Lightroom Classic 2024 and trying to create a panorama with images that have been rotated slightly using the crop tool because the images I want merged were all tilted at different angles apparently. However, when I do a panorama merge, the images are sticked together as if the angle correction did not take place. I attached a sample of what my panorama looks like right now. It is approximately a 150 degree panorama and obviously, the lake horizon line is not supposed to be warped upwards where the mountains are. I know lightroom does non-destructive edits on photos, but since all my images are RAWs, is there any way to force it to take stuff like rotation changes into account when stiching panoramas? Thanks!

The only way to do this in Lightroom would be to export the cropped images as TIFF, import these TIFFs and merge them. If you use Photoshop merge (not Camera Raw) then you will do exactly that, because Photoshop cannot work directly with raw files. It gets RGB rendered images from Camera Raw and merges these.

I have the following trouble while merging several images in a panorama in Lightroom. Since some of my images are from some unknown reason blurred at the edge I tried to crop the blurry part out and merge them into panorama after cropping. However, the resulting panorama has always used the original size of the pictures resulting in an image where focused and unfocused sections are continuously changing. Is there some way how to merge only cropped images into a panorama without using other software?

You could try it by exporting the cropped versions as tiffs and then do the pano from those. The cropped tiffs will have to have enough detail overlap from one image to another for it to work. 25%-30% overlap should work ok.

Well, that is exactly the extra step I would like to evade. Ideally, I would like to crop the picture and merge it straight away and still have enough possibilities to adjust all raw related settings.

The problem with this approach is that Lightroom does not always take the sharper overlapping part. From my last experience I may say that if I have picture A on left and B on right then Lightroom prefers left part of picture B over right part of picture A even though the other way around would provide better result.

I never used the stitcher calss, but I think that you may get the estimated homography matrix at each pair of images, if you could obtain it easily, then you can multiply it with the corners of the first original image and so for the corner of the last original one, you will get their stitched coordinate, then get the min of left and right x-coordinates and min of up and bottom y-coordinates of each images. You may get the coordinates of of each stitched image, what you need to do in some cases of cropping.

I have a 360x180 panorama photo. I took the photo with an app (Google PhotoSphere). Unfortunately my app doesn't save the individual photos I took, only the auto stitched panorama, with the "center" of the photo of its choosing.