- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi, there
I am using SR300 to develop an AR application in which we need to scan user's face and to generate its mesh. But I found that in SDK 2.0 we are only able to get real-time 3D point cloud.
Though with point cloud it is practical to develop a fushion algorithm, but why bother? As far as I know, the old versions of realsense SDK enabled developers to get mesh from scan processing
directly. I was wondering that if I can achieve my goal directly in sdk2.0. And if the answer is no, where do I find the resource of old versions of SDK and samples? So far I failed to found the old
versions of RealSense SDK.
Link Copied
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I do not know of any plans to provide a new 3D object scanning program in SDK 2.0. Features that were in the old Windows SDKs are more likely to be achieved in 2.0 by using it with OpenCV and / or ROS.
For example, OpenCV can use a system called Structured Light to generate a 3D model. Google 'structured light opencv' for more details. Also, here's a video of the process.
https://www.youtube.com/watch?v=O5P65CYqo_Q GSoC 2015 OpenCV Structured Light - YouTube
The oldest version of the SDK that Intel recommends for use is the '2016 R2' version. It can be downloaded in your browser as a 1.8 gb download that can be launched immediately with the link below. R2 includes the 3DScan sample program that enables the creation of a 3D model and its export as an .obj format 3D model file.
http://registrationcenter-download.intel.com/akdlm/irc_nas/vcp/9078/intel_rs_sdk_offline_package_10.0.26.0396.exe http://registrationcenter-download.intel.com/akdlm/irc_nas/vcp/9078/intel_rs_sdk_offline_package_10.0.26.0396.exe
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page