GPU-accelerated descriptor extraction process for 3D registration in Augmented Reality
Timothy Garrett, Rafael Radkowski, Jeremy W. Sheaffer · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · 2016
Augmented Reality (AR) is a type of human-computer interaction that overlays virtual information on a user's natural visual perception of the environment. Determining where to place these virtual objects so they appear a part of the environment requires real-time tracking. One type of tracking uses point clouds, processed from commodity depth sensors, such as the Microsoft Kinect. In this format, tracking becomes a 3D registration problem which can be quickly solved using local registration methods; however, these techniques require a sufficient overlap between the object-to-track, and the sampled object points from the range camera. 3D descriptors can be used to provide this initial overlap, but require significant computational resources. This work extends previous research towards real-time descriptor computation on large point sets. To do so, the 3D descriptor extraction process is broken down into fundamental algorithmic steps. These steps are optimized by computing them on the GPU using NVIDIA's CUDA framework. Results of an experiment indicate the opportunity for large speed-ups in the most computationally intensive portions of descriptor extraction.
Cited by 7 works. doi:10.1109/icpr.2016.7900108
Authors
References
- Multidimensional binary search trees used for associative searching
- The Elements of Statistical Learning
- A Survey of Augmented Reality
- Object modelling by registration of multiple range images
- Iterative point matching for registration of free-form curves and surfaces
- 3D is here: Point Cloud Library (PCL)
- KinectFusion: Real-time dense surface mapping and tracking
- Fast Point Feature Histograms (FPFH) for 3D registration
- Scalable Parallel Programming with CUDA
- Multiview registration for large data sets