(Ground) Control Points and Alignment

One smallish project I tested did not allign all images into one component but 2, even though there were some very distinct features (handwritten numbers) present on images in both components. I tried adding control points to help with the alignment. Maybe I did not distribute them enough over the surface, which is due to a texture with few (human) recognisable features. After doing that, the 2 ccomponents were merged into one, but now there were faulty parts in the mesh. My guess is, that the control points were too close together in too small an area, so that the natural placement errors multiplied and sreweed up the alignment. I could swear I saw an option somewhere that said something to the extent of “control point priority” (which I thought meant that I can define, whether control points are absolute or just helpers for alignment), but I cannot find it again for the life of me. Did I dream that??? :slight_smile:

And a follow-up question relating to that:
When I have defined sufficient control points and hit alignment again - is that a completely new process or just a refinement? That is interesting in terms of processing times.

Another follow-up:
I don’t quite understand the components and why they remain after alignment - do they work similar to the chunks in photoscan? Does that mean that the second round of alignment is rather equivalent to aligning chunks?

Hi Götz

One smallish project I tested did not allign all images into one component but 2, even though there were some very distinct features (handwritten numbers) present on images in both components. I tried adding control points to help with the alignment. Maybe I did not distribute them enough over the surface, which is due to a texture with few (human) recognisable features. After doing that, the 2 ccomponents were merged into one, but now there were faulty parts in the mesh. My guess is, that the control points were too close together in too small an area, so that the natural placement errors multiplied and sreweed up the alignment. I could swear I saw an option somewhere that said something to the extent of “control point priority” (which I thought meant that I can define, whether control points are absolute or just helpers for alignment), but I cannot find it again for the life of me. Did I dream that???

Yes you guess right, its best to set the CPs a bit spread over the image
take a look how i do it https://www.youtube.com/watch?v=d8naLEtLqDY

Have seen it good, not dreamed about it :smiley: There is a option to set higher “priority” or better WEIGHT. Just click on CP and get a small info on the point and there can set it weight of the ControlPoint

And a follow-up question relating to that:
When I have defined sufficient control points and hit alignment again - is that a completely new process or just a refinement? That is interesting in terms of processing times.

  1. if have 2 or more components, and add CPs then need align it. But as have seen our alignment speed is quite fast even with 1000´s of images.
  2. When get bad alignment even with the CPs try delete all components in the project and align it again.

I don’t quite understand the components and why they remain after alignment - do they work similar to the chunks in photoscan? Does that mean that the second round of alignment is rather equivalent to aligning chunks?
Sometimes results are separated to component. but we use the idea much deeper and mostly for large datasets.
Will prepare a indepth explanation how to work with Components and how to use to its maximum potential. its a very powerfull tool and easy to manage, but it can look complicated on first look.

Wishgranter wrote:

Will prepare a indepth explanation how to work with Components and how to use to its maximum potential. its a very powerfull tool and easy to manage, but it can look complicated on first look.

Looking forward to this!

Here is one quick tutorial video how to merge two components using 4 control points

https://www.capturingreality.com/video/tutorials/cp-join.wmv

Wishgranter, thanks, I am relieved now! :slight_smile:
How exactly does this weight work then.
Is it relative between the different CPs or is it absolute in relation to tie points?
Can you describe a scenario with a couple different settings and what this changes?

Thanks for the video, martinb!
As I see it, you really only use the CPs as helpers since you don’t spend ages placing them.
Or is that just my lack in experience? :slight_smile:
Is it correct to assume then, that they do not have a higher priority than tie points?
What I am aiming at is: can falsely placed CPs screw up the automatic alignment?

Thanks for posting the video martinb.
Really cool software! :sunglasses:

Götz Echtenacher wrote:

Wishgranter, thanks, I am relieved now! :slight_smile:
How exactly does this weight work then.
Is it relative between the different CPs or is it absolute in relation to tie points?
Can you describe a scenario with a couple different settings and what this changes?

Thanks for the video, martinb!
As I see it, you really only use the CPs as helpers since you don’t spend ages placing them.
Or is that just my lack in experience? :slight_smile:
Is it correct to assume then, that they do not have a higher priority than tie points?
What I am aiming at is: can falsely placed CPs screw up the automatic alignment?

Is it relative between the different CPs or is it absolute in relation to tie points?
= can set diff settings on weight per CP
= as get higher weight on the CP it can “tear” the images with the CP out of alignment
try with default settings, its work, when not then is better to add CP to more images as setting higher numbers on weight

Hi martinb,
There are some things I can’t figure out that you are doing in the video. Would it be possible for you, or Wishgranter, to put together a PDF that explains the steps you are taking in the video?
Thanks,
Bill

Hi Bill
Or can guide you trough the process over remote access software co can see it in action and get it the rigth way ?
if yes PM me on muzeumhb@gmail.com

Thanks Wishgranter,
I’m starting to get the hang of it. I want to get bit more familiar with it before I take you up on your offer. In the meantime, I have joined 7 components using 8 control points. I noticed in the information panel for the control point there are inputs for Actual Position. I know the coordinates for some of the points. Is it possible to enter the actual coordinates for the control points I know, and disable the control points I don’t know the coordinates for, to register the model?
Bill

Was able to add Ground control, update the project, and export a mesh in the correct coordinate system. Let me know if this is a correct procedure or if there is something I should have/could have done differently.

Hi Bill

you want combine the Components together or need Geo-reference the project ?

Wishgranter wrote:

Hi Bill

you want combine the Components together or need Geo-reference the project ?

Hi Wishgranter,
Here is what I did after I initially ran Align Images. I visually inspected the components and deleted the components that were not useful. I then added 4 control points to 2 photos in each of the remaining components (Component 0 through 9 using Control points 0 through 3). I then ran Align Images again. This resulted in Component 0 (1) and Component 3. I created 4 new Control points, point 4 through 8, and assigned those to Component 0 (1) and Component 3 and ran Align Images again. This resulted in Component 0 (2). I created 5 new Control points, point 9 through 13, and change the type to Ground control and added coordinates. These were assigned to 5 images in Component 0 (2). I then unchecked Control points 0 through 8 (type Tie point) and ran Update under the ALIGNMENT tab. It appears to have achieved the results I was after, but may not have been the proper or most efficient route to get there.

Update (after more playing around):

Hi Bill,

I think what you meant in the video was the cycling through the cameras, right?
That seems like a very handy function - it works with left and right arrows… :slight_smile:

About the alignment “screw-up”: Is there a setting that I can use if I only want to tell RC where to look for automatic matches but otherwise discard my control points? Is the standard setting (which I did not touch btw) already that? Maybe I am looking at it the wrong way but to me it would make more sence that way in cases of some minor alignment difficulties and it would give me the peace of mind that I did not make things worse since the algorythms seem to be doing a great job.

Hi Götz

the best way for CPs is to set in once on all components that want align, prefer to work with the components with highest img count, as our algo will try to align them together with all remaining components (imgs)

have say 5 components, will try to add CPs to all components, then align, if get 2 components, will add CPs to additional imgs…

ad if set CPs use ALIGN not UPDATE.

for the CPs and Components need prepare a longer tutorial. So its more easier to see it in action…

Hi Wishgranter,

I guess my problem in thinking about it results from a different project - the complex abbey choir ruin.
There I have about 750 images and the alignment took the whole night.
It resulted in one component all right but has some falsely aligned areas (or groups of cameras) which resulted in two surfaces in a short distance from each other.

Firstly it was quite a challenge to figure out which cameras are affected (but manageable thanks to the handy inspection feature). And then I am not sure what will happen with the CPs or rather when to stop with them since I will always have some images or cameras where my CPs will not cover the whole are but only say half.

Anyway, I will try out some more and am looking forward to the tutorial.