Messages

136

Face and Body Scanning / Re: Face scanner - Prototype

« on: December 15, 2013, 03:03:45 PM »

Not yet  , ill try to fix the illumination source so it doesnt move relative to the subject, that may increase geometric reconstruction quality.

Also ill try Meshlab's noise filtering options.
 

137

Face and Body Scanning / Re: Face scanner - Prototype

« on: December 13, 2013, 05:34:13 PM »

Results are still too noisy... i need to sort that out.
I dont like Zbrush smoothing brush, a filter as someone had mentioned here would be of course better, (Noise should of course be sorted out by increasing the scans quality...)

---

Im getting like 45k points, is that too few ?
Maybe the quality between correspondances is not good enough ? what numbers should i aim for ?

Thank you in advance for your help!

138

Face and Body Scanning / Re: Face scanner - Prototype

« on: December 13, 2013, 02:15:28 AM »

Okay, Ill post a few results (spraying paint on the subjects face) as well as texture rendering screenshot.

139

Face and Body Scanning / Re: Face scanner - Prototype

« on: December 12, 2013, 09:40:55 PM »

Mesh is still a little bit noisy, i guess i need more corresponding points to be found, at the moment i have around 39k.

140

Face and Body Scanning / Re: Face scanner - Prototype

« on: December 12, 2013, 09:38:05 PM »

I promised some examples of my system, so here it goes :
(Those are very early tests, everything will be improved with noise projection)
Tell me what you think!

141

Face and Body Scanning / Re: Face scanner - Prototype

« on: December 11, 2013, 08:36:53 PM »

Emage, Mr_Curious Thank you!
I hope ill be able to post some tests tomorrow as well as some issues for sure! I just bought my photoscan license finally

Alexey, what do you think of this approach ? could you tell me any issues that the software could have with this system ?

142

Face and Body Scanning / Face scanner - Prototype

« on: December 11, 2013, 01:43:28 PM »

Dear all,

After many questions, doubts solved and problems addressed - Im very happy to share with you all my very first prototype of Stereophotogrammetry face scanner: 

- https://www.dropbox.com/sc/suy6fwejacnbn50/aSMR2CZQ4E -

--------------- Key features: -------------

* I really wanted it to be "cheap" although at the cost of not being totally instant (it takes 2 seconds to go from 40 degrees to 140 degrees (there are angle limitations for the body, security of the cameras etc)

* It is important that the subject doesnt move - for alignment purposes.

* The scanning is done with total darkness as of yet, every camera opens their shutter for 1/50, in between that time, the flashes are fired at 1/16intensity.

* The prototype has a inclinometer (acelerometer) sensor on the rotation axis (similar to the ones our mobile phones have to detect inclination. (its the blue box)

* When the inclinometer detects the programmed ? angle it sends the shutter signal to each of the cameras and flashes

---------- Details -----------

Pre-mirror up signals are sent so the camera's total shutter lag is decreased.

- Time between fully shutter mechanical operations is around 0,4 seconds. That is : The camera is not ready to shoot again until 0,4 seconds.
- Cameras are Canon 700d (i bought those ones because they had better FPS)
- Flashes are Yongnuo 560 iii
- Lenses are canon 1.8f 50mm
- The radius of the semi circumference is 80cm
- Distance between cameras is around 26cm

* For 6 programmed angles, the total shooting time is 2 seconds of scanning.
(shoots every 20 degrees) 6 times. That is : 36 photos. (Maybe i?ll add two more cameras), but no more.


------------------------------------- ISSUES ------------------------------------------------


In my opinion, the fact that the flashes are moving relative to the subject is not very good for feature detection

Subject movement is a problem of course, but if i do the movement within 2 seconds and the subject tries to stand still, alignment is done correctly.


-----------


I Would really love to read opinions, thoughts or critics of course! 
* I?ll keep this post up to date with any new tests, problems or results so that we can all share our experiences!

https://www.dropbox.com/sc/suy6fwejacnbn50/aSMR2CZQ4E

143

Face and Body Scanning / Re: Type of pattern projection

« on: December 10, 2013, 04:28:39 PM »

Thank you a lot Mr_Curious, this lot of information is more than good enough for me to start moving things.
I?ll probably start buying two strobe light projectors and see the results. (My budget is not very high) (And im just scanning 180? of the face at the moment, not 360? head)
- What i meant with the loose of the diffuse illumination was not for the texture, but for the feature point calculation (to have some parts of the head in shadow because of the projector flash)

I?ll keep you informed.

xManuel.

144

Face and Body Scanning / Re: Type of pattern projection

« on: December 10, 2013, 04:00:34 PM »

I?ll try it this week and post results
Do you think the results (Improvements in the n? of points detected) is worth the money investment ?
Do you use more than one strobe light projector for the sides of the head ? or just one ?

- I?ve been thinking on using it because some girls have a really smooth skin, this would cause problems.

Thank you!

145

Face and Body Scanning / Re: Type of pattern projection

« on: December 10, 2013, 03:50:15 PM »

However, this will kill my diffuse illumination set up, is it worth it ?
A really high lumen projector usually costs more. Isnt it better to use a strobe light projection ? with a flash attached to a 50mm lens for example ? (or even wider) (and a slide ofc)

- From what i?ve read on the forum, strobe light projection have more advantages to DLP projection method.

Thank you for your kind responses!
Regards,
Manuel.

146

Face and Body Scanning / Re: Type of pattern projection

« on: December 10, 2013, 01:35:50 PM »

The problem with the DLP projector is that its light doesnt give me enough intensity (to have a good contrast ratio) in comparison to the exposure of the flashes.

147

Face and Body Scanning / Type of pattern projection

« on: December 09, 2013, 06:57:26 PM »

Hello all,

I?ve finally made some improvements on my 3d scanning rig, the next step for me (at the same time as improving light conditions) is to develop a pattern projection.

So my doubts are the following:

1. What kind of noise projection should i use ? I know that SIFT algorithm tries to find "corners" so as little as possible squares would be very appropriate. But maybe noise(dots) pattern works best ? anyone had any experience with this ?

2. Im going to use a flash with a 35mm transparent slide and a lens to project the details onto the surface, however using this method i?ll loose my diffuse light conditions...
The problems are that:

- The projection has to be much more powerful than the default light conditions so i can achieve contrast... otherwise noise projection would be useless i suposse.
- I?ll have to lower the exposure of my flashes so the noise pattern projection flash fires with more intensity in comparison to the other flashes (this however, will kill my diffuse light)

How could i overcome this ?

- I?ve read that i could use laser with different patterns however this is dangerous for the eyes
- IR not an option ?
- UV maybe ?

Thank you for your always kind responses!
Manuel.

148

Face and Body Scanning / Re: Face scanning and macro

« on: December 03, 2013, 06:39:10 PM »

Im using Canon 700d (i didn?t buy canon 600d because they have less FPS / time between shutters is larger)

- Thats not a problem as of now. However i understand what you mean.

149

Face and Body Scanning / Face scanning and macro

« on: December 03, 2013, 01:40:36 PM »

I have some questions regarding to face scanning id love to discuss / share some thoughts about this with all of you:

- Ive recently made a first prototype with good scanning results, however it is too big at the moment. (it is a semi circular frame of 80 cm radius) with 6 cameras mounted onto it.
This is the source of my problem, i need this system to be smaller. So here go all my questions:

1. At the moment the cameras are positioned at 80cm from the subject so i get "proper" DOF (its around 17cm) lenses are 50mm prime. f13. Im using continuos light using high luminance fluorescents.

2. Since i need this system to be smaller, (like 50 cm radius for example) DOF will of course decrease.
I can probably stop down to f20 but difraction will blur my entire scan producing a low amount of matching points -> more noise.

3. 50mm lens allows me to focus at a minimun distance of 45cm so 50cm is OK, however this is not the main issue.

4. Ive calculated 80cm (which is the radius of this prototype) it is so to fill the frame of the sensor.
If i build a new one at 50cm the face will fill too much of the face leaving un-scanned parts. So i may have to change to 35mm lens.

5. The main problem i see here is the DOF, does anybody have experience / thoughts / ideas scanning with macro photography ? How could i overcome DOF ?

* Maybe playing with cameras focusing different parts of the face ?
* Maybe increasing the number of cameras focused at different / overlaping parts of the face ?

- Ill post a very quick preview of the 3d render i did before the prototype so you can understand better what im saying:

All ideas or thoughts are welcome!

Manuel Franquelo.

150

General / Re: Different Lighting source for feature detection

« on: November 18, 2013, 05:55:24 PM »

Hello Alex,
Thank you a lot for your response.

What about if the light source is diffuse ? Light direction would change of course, but the change would be very subtle.
Is the feature detection affected really negatively by this approach ?