Forum

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.


Messages - Andrew

Pages: 1 [2] 3 4 ... 6
16
General / Re: TUTORIAL - Request
« on: April 16, 2014, 12:08:01 AM »
I personally don't import camera positions, it's just too easy for fixed cameras (and their lenses) to get gradually off their previously calibrated positions. Also, if you are using multicam rig with good lighting and some of your cams fail to align, it _usually_ means your photos are not sharp enough, or your cams are spread out too far etc. In case of really pale skin/featureless subject, you can go to 'Tools/Export/Export_Cameras', save your 'sphere' camera setup and then load it ('Tools Import\Import_Cameras' when you switch to your 'face' setup.

-Andrew

17
Python Scripting / Python scripts collection?
« on: April 13, 2014, 09:30:33 PM »
I use Photoscan Standard edition - looking at feature list (considering my needs) I just couldn't find justification for upgrade. When looking at "Python scripting support" line I assumed it meant basically a more advanced and open alternative to "Batch Process" feature, but every now and then I read it can do lots more, hence my question: are any useful scripts included in Pro demo installation package? Or maybe there is some repository of useful scripts available? Or at least a list of what some more advanced scripts can accomplish? I think more insight into this can help push me and many other over the edge to upgrade.

-Andrew

18
General / Re: A Few Questions Graphics Card and Camera Lens related
« on: April 12, 2014, 07:25:01 PM »
I too had pretty bad experience with Nikon D3200 - it produced VERY soft images.
One thing to take into consideration - D3200 is actually last year's model, now replaced by D3300 with low pass filter removed, much like D7100 that I use and am very satisfied with.

However, I can't say that 24mpixel worth of OLP-less imagery is noticeably better than 600D's. They are really, really close, which puts Canon at clear advantage - less pixels means less computations and less memory to produce virtually identical scans.

-Andrew

19
That's very interesting indeed, Lee! None of us expected any improvement unless the source (cameras) were USB 3.0 as well.

From what I gather, you are using 3.0 hubs, 3.0 cables between hubs and PC, and obviously 3.0 in your PC. But, did you keep the 2.0 miniUSB->USB from cameras to hubs? Do these even exist in 3.0 flavor? I did a search in some local stores around here and couldn't find any.

-Andrew

20
General / Re: Accuracy of masking from low/med quality model
« on: April 05, 2014, 10:33:05 AM »

- where do these imprecisions in masking from model come from? Could it be that masks are generated for undistorted photos but the resulting mask is displayed over original photo that still carries optical distortions? In this case, mask could actually be perfect even if it appears to be off.

- does Photoscan respect photo masks down to pixel precision? Or is there some threshold that causes PS to scan for points even if they fall just slightly behind mask edge?


Alexey, could it be possible for you to shed some light in this matter? Masking from model seems like a very useful feature, but for now, it seems nobody fully understands how it works (judging by lack of answers).

-Andrew

21
For your face rig, if you haven't already try switching over to USB3. It's far more reliable, you can overcome distance issues associated with USB2 cables, you can send more data and hook up more cameras per PC.

Thanks for the tip, Lee! I actually do hook things up to USB3.0 on my Lenovo notebook, but since cameras themselves (600D's) are 2.0, I didn't bother to search for 3.0 cables and hubs. It didn't occur to me that it would make a difference, unless cameras were USB 3.0 as well...

-Andrew

22
General / Re: Accuracy of masking from low/med quality model
« on: April 01, 2014, 09:07:46 PM »
FLuca, to remove masks, right click on your chunk in Workspace pane, and click Remove/Remove Masks

23
General / Re: Accuracy of masking from low/med quality model
« on: March 31, 2014, 11:04:35 PM »
I often shoot in scenarios where hand masking would be too tedious (for instance building rooftop against sky, with occasional tree branches obstructing view). When shooting huge projects, I cant really afford excessive overlap (I risk running out of RAM) so I really need to rely on all available photo information, including what mask from model sometimes clips...

I am curious to hear what Alexey can tell us about causes of mask clipping and whether that is a problem at all :)

-Andrew

24
General / Accuracy of masking from low/med quality model
« on: March 31, 2014, 01:14:36 PM »
I read recommendations of masking photos from model on more than one occasion on this forum, but whenever I tried the workflow (align unmasked photos first, build low/med quality mesh and use it to generate masks for photos), there would be enough imprecisions in the mask to scare me away from using this feature. Mask encompassing some of the background is not the end of the world, but often mask would clip tip of the nose, or edge of ears etc.

And so, my questions are:

- where do these imprecisions in masking from model come from? Could it be that masks are generated for undistorted photos but the resulting mask is displayed over original photo that still carries optical distortions? In this case, mask could actually be perfect even if it appears to be off.

- does Photoscan respect photo masks down to pixel precision? Or is there some threshold that causes PS to scan for points even if they fall just slightly behind mask edge?

Cheers,
Andrew



25
Luckily the rooftops were sloped and could be photographed even from ground level, but I did shoot them from some distance, and with camera raised high on monopod (height of approximately 3,7m).
Drone would certainly help though :)

26
That's exactly what I meant, lifting your monopod over your head and triggering via wireless trigger in your hand. Depending on your height and how tall your monopod is, you can take photos from 3.5m, give or take. Of course you need to compensate for camera shake (you're NOT gonna be able to keep it steady), and framing what you don't see can be tricky.

27
So far we survive without drones but they are on the shopping list :) You can achieve quite a lot with tall monopod and wireless trigger.

As for overlap:

- I tend to shoot for too much overlap rather than too little,
- I try to be really aware of what each photo captures, memorize features and make sure they end up in next frame
- I am really methodical about moving along captured geometry. I try to shoot in rows, I tend not to get distracted with cavities (and such) that require additional photos, I get back to those later.
- I try not to shoot multiple camera angles from same position, it's better to just move your (camera's) position
 
Hope that helps!

28
Approach to scanning bigger things? In short - lots and lots of photos with enough overlap and with hope for lighting conditions not to change throughout :)
It sometimes helps to not only shoot entire thing from consistent distance that will give you target fidelity but also take some snaps from slightly larger distance - helps with aligning photos in problematic areas, but it can at times decrease accuracy of alignment - quite tricky.

-Andrew

29
Thanks!
Unreal4 holds a lot of promise, but without a sizeable team of programmers I wouldn't recommend taking the leap just yet (unless it just for visualizations with light scripting). We are experimenting with it a lot so that we can hit the ground running a couple of months from now, when it matures enough for us to port the game to UE4 (and target more platforms). Hopefully we will get to improve some scans in the process as well, a couple of them from the preview are almost two years old, I (and Photoscan) have come a long way since :)

Andrew

30
Hopefully you guys find it interesting what we're cooking up here using Photoscan. Some interactive scan samples included in the article.

http://www.theastronauts.com/2014/03/visual-revolution-vanishing-ethan-carter/

Cheers,
Andrew

Pages: 1 [2] 3 4 ... 6