2016 · 22 citations · 11 references
ComputingComputer Science3D ModelingComputer VisionMachine VisionDepth EstimationDepth PerceptionMotion AnalysisMotion CaptureMotion EstimationTracking SystemsColor Image ProcessingDepth SensingGround-truth ValidationLight Reflection PropertiesMotion TrackingMotion Tracking AnalysisMotion Tracking SystemsThree-dimensional Motion Capture
This paper presents an evaluation of the Tango tablet regarding its motion tracking and depth perception capabilities. A methodology for performing such kind of evaluation is proposed. Motion tracking error is assessed in both small workspace and large environments. In the small workspace scenario, the distances reported by the motion tracking system are compared with values measured with the aid of a graph paper. In the large environment condition, the tracking error consists in the difference between initial and final positions given by the system when the device moves around the environment and returns to the same location. Depth sensing precision is evaluated by comparing the 3D coordinates reported by the system of the inner corners of a chessboard pattern with ground truth values obtained from color camera image processing. The results show that Tango tablet sometimes presents large motion tracking errors, which may harm AR experience. In addition, Tango tablet depth sensing presents average error values similar to desktop depth cameras, but it is more sensitive to infrared reflection properties of the objects to be mapped.
11
Bilateral filtering for gray and color images
Carlo Tomasi, Roberto Manduchi · 2002 · 8.1K citations
Arts, Vision Science, Color Vision +16
KinectFusion: Real-time dense surface mapping and tracking
Richard A. Newcombe, Andrew Fitzgibbon, Shahram Izadi et al. · 2011 · 3.9K citations
A benchmark for the evaluation of RGB-D SLAM systems
Jrgen Sturm, Nikolas Engelhard, Felix Endres et al. · 2012 · 3.9K citations
Multiple view geometry in computer vision
H. Opower · Optics and Lasers in Engineering · 2002 · 3K citations