Forum

Author Topic: Advanced setting for dense reconstruction  (Read 6038 times)

dense_data

  • Newbie
  • *
  • Posts: 8
    • View Profile
Advanced setting for dense reconstruction
« on: July 08, 2018, 09:57:36 PM »
I've attempted to perform a rather ambitious dense reconstructions with PhotoScan Standard. While everything works quite fast because of GPU acceleration, I've hit the (apparently) well known bottleneck of dense point cloud create after successfully creating depth maps. There seem to be a couple tricks posted in this forum which should speed up the process namely by reducing the number of maximum neighbors, because "excessive" overlap seems to cause PhotoScan to become sluggish. Now the problem is *those* settings are apparently only accessible in the form of python commands, e.g. as shown here http://www.agisoft.com/forum/index.php?topic=8361.0

Is there any way to adjust these settings in PhotoScan Standard?

(Hint: This would also be nice as a new feature, GPU accelerated dense point cloud creation)

dense_data

  • Newbie
  • *
  • Posts: 8
    • View Profile
Re: Advanced setting for dense reconstruction
« Reply #1 on: July 12, 2018, 03:49:27 PM »
Can anybody help?

Alexey Pasumansky

  • Agisoft Technical Support
  • Hero Member
  • *****
  • Posts: 14813
    • View Profile
Re: Advanced setting for dense reconstruction
« Reply #2 on: July 12, 2018, 08:00:31 PM »
Hello dense_data,

Starting from PhotoScan version 1.4.2 (both Standard and Professional) you can modify that "hidden"setting using Tweaks dialog that is accessible from the Advanced Preferences tab.

Create the new entry with the following name:
Code: [Select]
main/dense_cloud_max_neighborsand assign the desired pairs value, like 50 or 60, for example.

To remove the limit you can either delete the "tweak" or set the value to "-1".
Best regards,
Alexey Pasumansky,
Agisoft LLC

dense_data

  • Newbie
  • *
  • Posts: 8
    • View Profile
Re: Advanced setting for dense reconstruction
« Reply #3 on: July 13, 2018, 12:05:48 AM »
Thank you, very much! Looks much better already!
« Last Edit: July 13, 2018, 12:11:25 AM by dense_data »