Known cameral positions

[Andrea Alessi]

It really depends what your final requirements are. 6 cameras are a very low amount, but for certain scenarios it can be enough. In that case, you would however need pro/aerial so that you can import safely a known camera configuration rather than re-computing the first step each time.

I'll be happy to give you a free aerial when you're ready.

No, converting a JPG to TIFF will not help. If you have raw images and need HDR texture then yes, you can save as tiff. But if not, then you can safely use JPGs.

[Karlsc]

would it help if I first converted the JPGs to a full resolution format like TIFF?

[Karlsc]

and the last two are too big.. I will send by email (2.9 and 3.2 mb)

[Karlsc]

sorry having to go 1 by one due to size

[Karlsc]

[Karlsc]

here are the images from the rig that are not working (upladed in sets of 3 because of size limits)

[Karlsc]

That's not necessary. unless it would help me with coming up with a minimum camera config for the rig. right now i'm at 6 cameras arranged 3 vertically spaced about 45cm apart with a plan divergence of about 10cm (intentional)

then I have two cameras at 45 degree angle to this about 30 cm fwd of the vertical plane... and aligned with the center of the vertical plane... then I have 1 camera shooting straight down about 1.5 meters above the main line of the cameras

that's my latest configuration see attached.. if I can get consistent results (if I need to add cameras please make suggestions) I would like to take you up on your offer to try Arial again fo see if importing the configuration in the wild works for me. Until I get funding I cannot justify the $2500 for a full license.

[Andrea Alessi]

If you absolutely need 100% determinism, then you should use only CPU. Determinism is not guaranteed on GPU due to floating point approximation - this is expected and known, but if you're not convinced i'll happily forward you some tech detail on why that happens.

[Karlsc]

my concern is that I am NOT getting deterministic results

[Karlsc]

I dunno what qualifies as bad photos. these are 4k images that are from static shots not video.

[Andrea Alessi]

I understand the question now, thank you for claryfing.

Yes, with the same settings you will obtain the same results. Zephyr is deterministic. (please note that if you set certain custom values and then switched to preset, zephyr will keep some of them, so make sure to reset the settings if this is the case).

However, due to floating point math approximation over GPU acceleration you may get slightly different results (which in turn, cause zephyr to fail) and this is usually due to bad photos. If you run it in CPU zephyr is 100% deterministic (albeit slower than in GPU),

[Karlsc]

But if I present the tool with the same 14 images that it extracted the first time, shouldn't it logically extract the same structure from them as it did the first time? IE. I'm looking at repeatability

[Andrea Alessi]

Hi Karlsc,

you need to use the "load photos from known parameters function" which is available in 3DF Zephyr Pro and Aerial only. This is not possible in zephyr free or lite.

[Karlsc]

how do I get the system to reliably repeat recognizing the cameras for exactly the same images?