331
General / Re: Graphics cards and memory
« on: April 27, 2012, 06:55:03 PM »
Are these test results related to processing with Agisoft?
Show Posts
This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.
332
General / Re: Graphics cards and memory
« on: April 25, 2012, 05:12:50 PM »Just wanting to check if anyone have run any tests with the 680, or even compared it to the AMD?
I'm also interested to know.
Also has anyone run any tests with dual cards? whether it be Quadro 5000/6000's or GTX580? does this do anything with Agisoft during Align stage?
333
General / Re: Choice of lens for shooting small objects
« on: April 25, 2012, 05:11:50 PM »I'll write a quick tutorial on focus stacking tomorrow. However it can be done without. Using a tripod and long shutter times/ high sensitivity captures can be made at the smallest aperture, ensuring large depth of field. Stay tuned, already looking forward to results...
In my naivety I have only just hard of focus stacking. Great idea! interested to see how you combine the data. http://en.wikipedia.org/wiki/Focus_stacking
Ops! just saw the long list of software applications for this purpose.
1) Interested to know if anyone has any experience with these? HeliconSoft looks interesting - http://www.heliconsoft.com/video.html
2) Also I wonder what impact these digital alterations will have on the alignment and build of images to meshes?
334
Feature Requests / Automatic Masking.
« on: April 25, 2012, 04:57:26 PM »
I have already spoken to Alexey about this idea. But wanted to get it out there for others to ponder.
Automasking. A technique of using a set of images empty and a set with your subject in, full. If you think about it you can Mask between the 2 automatically as a difference. With or without.
I hope we can see something like this introduced into Agisoft. Alexey seems interested in it but wonder what others would think of how it could be introduced into the pipeline? perhaps as a tick box and input line field to point to the sessions empty set of images folder to process with at Align or Build stage? This would be ideal for video sequences and head or body captures, when using allot of images.
The idea of then being able to apply this in batch on dozens of images during Align or Geometry build, without the need to mask by hand or use a Chroma Key background with colour spill is an appealing one! at the moment my captures suffer allot with background objects (blobs) getting introduced in the final build model if not masked.
You can of course do Difference Matte in Aftereffects or Difference filter in Photoshop by hand but the results are a little tricky to get right but the idea is sound and it is rather time consuming!
If implemented it would then just be a matter of common practice to always take an empty set of images at the start of a capture session
Interested in a discussion.
Automasking. A technique of using a set of images empty and a set with your subject in, full. If you think about it you can Mask between the 2 automatically as a difference. With or without.
I hope we can see something like this introduced into Agisoft. Alexey seems interested in it but wonder what others would think of how it could be introduced into the pipeline? perhaps as a tick box and input line field to point to the sessions empty set of images folder to process with at Align or Build stage? This would be ideal for video sequences and head or body captures, when using allot of images.
The idea of then being able to apply this in batch on dozens of images during Align or Geometry build, without the need to mask by hand or use a Chroma Key background with colour spill is an appealing one! at the moment my captures suffer allot with background objects (blobs) getting introduced in the final build model if not masked.
You can of course do Difference Matte in Aftereffects or Difference filter in Photoshop by hand but the results are a little tricky to get right but the idea is sound and it is rather time consuming!
If implemented it would then just be a matter of common practice to always take an empty set of images at the start of a capture session
Interested in a discussion.
335
Face and Body Scanning / Re: Scanning people: Tips needed
« on: April 25, 2012, 04:46:08 PM »Which cameras are you using? Are there maybe camera's without preview screens and excess funtionality one can buy cheaply?
I use Canon 550D's because of the great support Breeze Software offers for multi-camera tethering. Not sure what else you can use. Sony's 24fps new Cameras look interesting, Pentax? Nikon? but all come with limited multi-camera support. Unless you can get some one to write something and of course the Cameras really need triggering and data input ports.
336
Face and Body Scanning / Re: Scanning people: Tips needed
« on: April 25, 2012, 04:43:24 PM »
Hey Mark,
Great to see cool scans! I always get a buzz when I see cool human scans in colour! It would be good to see a grab of the grey scale mesh, to see how the scan came out. The colour looks great.
Mark and Harold: I still use Stereo pairs, with a combination of other Camera positions. I use Agisoft simply because of it's user friendly graphical interface, python support and other features. Plus the results are "volumetric" and not just from a single calibrated pair of Cameras (oh and the fact you don't need waste time calibrating) The results you can get are really incredible. You just need well placed, well lit and well focused images.
Hoping to see more Humans scans here!
Lee
Great to see cool scans! I always get a buzz when I see cool human scans in colour! It would be good to see a grab of the grey scale mesh, to see how the scan came out. The colour looks great.
Mark and Harold: I still use Stereo pairs, with a combination of other Camera positions. I use Agisoft simply because of it's user friendly graphical interface, python support and other features. Plus the results are "volumetric" and not just from a single calibrated pair of Cameras (oh and the fact you don't need waste time calibrating) The results you can get are really incredible. You just need well placed, well lit and well focused images.
Hoping to see more Humans scans here!
Lee
337
Feature Requests / Re: GoZ for PhotoScan
« on: April 18, 2012, 07:36:21 PM »
+++1 Great idea! I'm doing just that at the moment with .OBJ, having a Goz button in and out would be fantastic! Although not sure how easy it would be to get Goz out from ZB to PS?
338
Feature Requests / Re: Multi Image Selection by Viewport
« on: April 17, 2012, 09:56:51 PM »
Geert.. so it is! stupidly I didnt try it with the select tools, was still on Arrow! Ha, thanks : )
339
Feature Requests / Multi Image Selection by Viewport
« on: April 17, 2012, 03:59:38 PM »
I was wondering if it might be possible to have interactive (multi) image selection via the viewport using selection tools?
It would make it easier to pin point single and groups of images/cameras by their alignment and position in the world, rather than sifting through the 2d image layout going through names of cameras.
Just an idea to prove the already incredible tool set.
Thanks, Lee
It would make it easier to pin point single and groups of images/cameras by their alignment and position in the world, rather than sifting through the 2d image layout going through names of cameras.
Just an idea to prove the already incredible tool set.
Thanks, Lee
340
General / Re: Agisoft PhotoScan 0.8.5 pre-release
« on: April 15, 2012, 04:03:46 PM »
Thanks for the new features in the new build Alexey and team, and for your hard work on delivering consistent updates.
341
General / Re: quantity vs quality of photos
« on: April 15, 2012, 03:56:32 PM »
To answer your question I would say:
a) more pictures with less resolution
b) less pictures with more resolution?
*c) more pictures with more resolution*
But if that isn't possible:
a) more pictures with less resolution
..but nothing lower than 10-12MP really.
the more you have of everything the better the data will be but of course this has an impact on processing time.
*c) more pictures with more resolution*
But if that isn't possible:
a) more pictures with less resolution
..but nothing lower than 10-12MP really.
the more you have of everything the better the data will be but of course this has an impact on processing time.
342
Feature Requests / Re: Smooth Mesh
« on: April 15, 2012, 03:49:29 PM »
I have to agree, although this is really a post processing problem. But I don't know of any decent tools in other applications, Meshlab, Mudbox etc that are able to Smooth a mesh but still retain some of the crisp details well.
Ideally a solution to smooth high frequency details but still retain the low frequency bulk volume would be ideal. Not sure how feasible this is though.
So kind of a +1 on this.
Ideally a solution to smooth high frequency details but still retain the low frequency bulk volume would be ideal. Not sure how feasible this is though.
So kind of a +1 on this.
343
Face and Body Scanning / Re: Scanning people: Tips needed
« on: April 15, 2012, 03:38:01 PM »
Really cool results! thanks for sharing.
It's tricky trying to build a multi-camera rig that can capture any kind of pose volume, so this handheld solution is ideal for people lying down or sitting.
The problems are in the focusing, triggering and exposure settings (minimizing DOF also). Capturing a single moment in time from multiple cameras is very hard to do, consistently! as you need multiple PC's networked and lots of patience.
I'm currently finalizing a full body rig at the moment
Did you think of trying a fig rig with a stereo pair of cameras? tethered and triggered by a PocketWizard mounted on the Figrig? This can also be quite useful.
It's tricky trying to build a multi-camera rig that can capture any kind of pose volume, so this handheld solution is ideal for people lying down or sitting.
The problems are in the focusing, triggering and exposure settings (minimizing DOF also). Capturing a single moment in time from multiple cameras is very hard to do, consistently! as you need multiple PC's networked and lots of patience.
I'm currently finalizing a full body rig at the moment
Did you think of trying a fig rig with a stereo pair of cameras? tethered and triggered by a PocketWizard mounted on the Figrig? This can also be quite useful.
344
Bug Reports / Build Quality inconsistencies across PC's
« on: March 05, 2012, 07:46:02 PM »
I have noticed an odd bug. Using 3 PC's I have tested using the latest build of Agisoft Pro.
PC01, PC02 and PC03.
The 5 Image/Camera file was Aligned on PC01 using the below settings.
On PC02 and PC03 (different CPU, RAM and GPU configurations) I found different build qualities. Some odd stair stepping effects, not present on the PC01 build.
-----------------------------> -----------------------------> -----------------------------> ----------------------------->
PC01
PC02
PC03
Each build is different. Is there a reason for this? can CPU or GPU differences affect the build quality?
If you would like a file to test, I can upload.
PC01, PC02 and PC03.
The 5 Image/Camera file was Aligned on PC01 using the below settings.
On PC02 and PC03 (different CPU, RAM and GPU configurations) I found different build qualities. Some odd stair stepping effects, not present on the PC01 build.
-----------------------------> -----------------------------> -----------------------------> ----------------------------->
PC01
PC02
PC03
Each build is different. Is there a reason for this? can CPU or GPU differences affect the build quality?
If you would like a file to test, I can upload.
345
Feature Requests / Re: Cull/Cut by Low-Density tool.
« on: March 05, 2012, 04:51:30 PM »Hello,
If you mean only a selection tool (like connected components selection for example) that selects faces based on their sizes, I think we could add such a tool to one of the upcoming updates, but deleting these faces will lead to the saw edged model (and it will be more visible in case of Height Field models).
I think it could possibly be quite useful, select by edge size similar to meshlab, which has a default setting like 0.01 +/- so you can choose, or similar to your % slider? The saw edge is no problem really, it is almost the same effect than if you process as Sharp. It's just Smooth seems to create a more useful mesh on the internal important areas, apart from the unwanted garbage areas which are good to remove if the user chooses to.
