Intro
After Dive 8 in which I had Lauren literally lay down on the Yukon in order to get a point of reference for the photos for the model, I was ready to just give up when that didn’t work. I’m not sure what happened with Dive 9 but I think it was so disappointing and fruitless that I didn’t bother publishing an update. I know that the dive was in May 2023.
I recently conducted Dive 10 with some friends (you know who you are!) with a special mission a couple weeks ago.
This is really getting old, but at this point I’m determined to finish my model of the Yukon — in spite of the “sunk cost” concept of accounting!
Dive 10 Results
Below is the state of the model before the dive. It is a “low quality” model but you can see that the areas on the starboard side are still missing.
The goal was to (obviously) fill in those blank areas.
The dive went pretty well and I got pictures of all of those blank areas; however, when I first added them to the model, I had yet another problem even though all the photos aligned:
Once again, you can see that the photos align with each other but not with the rest of the aligned photos. I think the reason for this is that I went long distances along a flat surface and even small errors can throw things off over those distances. You can see how it starts to slightly “lift away” from the main model right around the area of the stern gun.
Here is another view of that:
Dive 10 Only
In order to get some validation on JUST photos from dive 10, I built a model only using photos from that dive. From the “top” it looks good:
HOWEVER, if you look at the sparse (point) cloud from the “side” then you can tell this isn’t correct:
Combined Model
I grafted on the photos that I felt aligned pretty well with the existing model and got something which is fairly good but is still missing one section of the starboard hull. I did about six different models trying to get it to work.
Here is the last one I built that I feel balances correct alignment with “filling in the blanks”:
The area in the red rectangle really needs to be re-done and re-aligned. It is low quality right now so the overall model will look better once it is done; however, the alignment is off near the edges of the red rectangle and there is obviously a missing piece in the middle.
Here is a closeup of that area in the rectangle:
In this closeup view, you can really start to see the alignment problems, especially around the area to the right. You can ignore the “light color fringes” around the areas. That will be resolved with a higher quality model.
I’m not going to bother publishing the model on-line at this point.
Next Steps
I’m hoping to do ONE MORE dive on the wreck in an attempt to get the alignment right. This project has been a lot more of a challenge than I thought (or that it should realistically be). It is my nemesis.
This next time, I’m hoping to go “up and over” the wreck by starting photos on the deck area and then up and over the hull. I’m hoping that this will help align the photos correctly. It is a pain to do it that way for multiple reasons, but I think I just need to go do it.
