Elastic Fragments for Dense Scene Reconstruction
Authors:
- Qian-Yi ZhouQZQian-Yi Zhou
- Stephen MillerSMStephen Miller
- Vladlen KoltunVKVladlen Koltun
- 1 December 2013
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 15505499,p. 473-480
Abstract
We present an approach to reconstruction of detailed scene geometry from range video. Range data produced by commodity handheld cameras suffers from high-frequency errors and low-frequency distortion. Our approach deals with both sources of error by reconstructing locally smooth scene fragments and letting these fragments deform in order to align to each other. We develop a volumetric registration formulation that leverages the smoothness of the deformation to make optimization practical for large scenes. Experimental results demonstrate that our approach substantially increases the fidelity of complex scene geometry reconstructed with commodity handheld cameras.Keywords
This publication has 29 references indexed in Scilit:
- Dense scene reconstruction with points of interestACM Transactions on Graphics, 2013
- Robust real-time visual odometry for dense RGB-D mapping2013
- Modeling Kinect Sensor Noise for Improved 3D Reconstruction and Tracking2012
- Accuracy and Resolution of Kinect Depth Data for Indoor Mapping ApplicationsSensors, 2012
- DTAM: Dense tracking and mapping in real-time2011
- KinectFusion: Real-time dense surface mapping and tracking2011
- Point Set Registration: Coherent Point DriftIEEE Transactions on Pattern Analysis and Machine Intelligence, 2010
- Simultaneous Nonrigid Registration of Multiple Point Sets and Atlas ConstructionIEEE Transactions on Pattern Analysis and Machine Intelligence, 2008
- Parallel Tracking and Mapping for Small AR Workspaces2007
- Non-rigid registration using distance functionsComputer Vision and Image Understanding, 2003