You mentioned sea urchin, if it's black I think I know why.
It is possible that this is the result of multiple influences.
First of all, of course, this kind of needle-like structure object itself is very difficult to create.-This means that this thin structure only has a small number of pixels involved in the calculation (and "栗" has many photos involved so it succeeded)
After that, because it is black. (The premise is that I guessed right) resulting in subtle differences in light and shade and color changes
This makes it hypothetical to have the same 500 photos of sea urchins.
For example: sea urchin needles do not coincide with the body of the sea urchin, they will be seen by the software to participate in the calculation.
But they overlap at the next angle. Then the software will think that the overlapping part of the needle is an "outsider", and ignore the calculation.
The reason why the overlapped sea urchin needles are regarded as "outsiders" and ignored in the calculation is that the difference in light and shade and color change is very small, so that the software cannot see the sea urchin needles when the sea urchin needles and sea urchin bodies overlap.
Of course these guesses are based on the fact that the sea urchin you scanned was black.
How do you evenly take 500 photos of such a small target? Moving the camera or moving objects? How to ensure a high success rate of reconstruction after moving?
12 comments
This article is important. Taught.
scan will be easy for small target. Precise ,efficient.
@SibeYu Thank you for your description. I am not sure how the software works.....but anyway I make 3D models by the below method:
begin3d.com/en_uc_wild-organisms
kyushu-u.ac.jp/en/researches/view/240
You mentioned sea urchin, if it's black I think I know why.
It is possible that this is the result of multiple influences.
First of all, of course, this kind of needle-like structure object itself is very difficult to create.-This means that this thin structure only has a small number of pixels involved in the calculation (and "栗" has many photos involved so it succeeded)
After that, because it is black. (The premise is that I guessed right) resulting in subtle differences in light and shade and color changes
This makes it hypothetical to have the same 500 photos of sea urchins.
For example: sea urchin needles do not coincide with the body of the sea urchin, they will be seen by the software to participate in the calculation.
But they overlap at the next angle. Then the software will think that the overlapping part of the needle is an "outsider", and ignore the calculation.
The reason why the overlapped sea urchin needles are regarded as "outsiders" and ignored in the calculation is that the difference in light and shade and color change is very small, so that the software cannot see the sea urchin needles when the sea urchin needles and sea urchin bodies overlap.
Of course these guesses are based on the fact that the sea urchin you scanned was black.
How do you evenly take 500 photos of such a small target? Moving the camera or moving objects? How to ensure a high success rate of reconstruction after moving?
@mathew4785 I like The Langoliers as well as The Shawshank Redemption.
Looks like a langolier.
@gastronomy 写真500枚撮りました。クリは一発でうまくいきましたが、ウニは何回挑戦しても失敗します。そこらへんの違いがよくわかりません。
栗のイガもここまできれいに再現できるんですね。
驚きました。
@alban Thanks, alban. I took 500 photos (the maximum of 3DF Zephyr Lite).
Amazing! How many pics?
@nebulousflynn Thanks, Thomas.
The fine detail here is blowing my mind
Staff Pick!