3D object reconstruction using multiple Kinect sensors and initial estimation of sensor parameters

In this paper, we reconstruct 3D object shape using multiple Kinect sensors. First, we capture RGB-D data from Kinect sensors and estimate intrinsic parameters of each Kinect sensor. Second, calibration procedure is utilized to provide an initial rough estimation of the sensor poses. Next, extrinsic parameters are estimated using an initial rigid transformation matrix in the Iterative Closest Point (ICP) algorithm. Finally, a fusion of calibrated data from Kinect sensors is performed. Experimental reconstruction results using Kinect V2 sensors are presented and analyzed in terms of the reconstruction accuracy.

[1]  Takeshi Masuda,et al.  Registration and Integration of Multiple Range Images by Matching Signed Distance Fields for Object Shape Modeling , 2002, Comput. Vis. Image Underst..

[2]  Akihiro Sugimoto,et al.  A Flexible Scene Representation for 3D Reconstruction Using an RGB-D Camera , 2013, 2013 IEEE International Conference on Computer Vision.

[3]  Vladlen Koltun,et al.  A Large Dataset of Object Scans , 2016, ArXiv.

[4]  Paul Checchin,et al.  Performance evaluation of 3D keypoint detectors for time-of-flight depth data , 2016, 2016 14th International Conference on Control, Automation, Robotics and Vision (ICARCV).

[5]  Vitaly Kober,et al.  Accurate generation of the 3D map of environment with a RGB-D camera , 2017, Optical Engineering + Applications.

[6]  Jiexin Pu,et al.  An Improved ICP Algorithm for Point Cloud Registration , 2010, 2010 International Conference on Computational and Information Sciences.

[7]  Bernt Schiele,et al.  3D Object Detection with Multiple Kinects , 2012, ECCV Workshops.

[8]  Vitaly Kober,et al.  Impulsive noise removal from color video with morphological filtering , 2017, Optical Engineering + Applications.

[9]  Vitaly Kober,et al.  Clustered impulse noise removal from color images with spatially connected rank filtering , 2016, Optical Engineering + Applications.

[10]  V. V. Kuznetsov,et al.  A method of face recognition using 3D facial surfaces , 2017 .

[11]  Hugo Jiménez-Hernández,et al.  Multiple Kinect V2 Calibration , 2016 .

[12]  Artyom Makovetskii,et al.  A modified iterative closest point algorithm for shape registration , 2016, Optical Engineering + Applications.

[13]  Marek Kowalski,et al.  Livescan3D: A Fast and Inexpensive 3D Data Acquisition System for Multiple Kinect v2 Sensors , 2015, 2015 International Conference on 3D Vision.

[14]  Rogério Schmidt Feris,et al.  Fine registration of 3D point clouds with iterative closest point using an RGB-D camera , 2013, 2013 IEEE International Symposium on Circuits and Systems (ISCAS2013).

[15]  Mitsuru Nakazawa Non-member,et al.  Calibration of multiple kinects with little overlap regions , 2015 .

[16]  Vitaly Kober,et al.  3D face recognition based on matching of facial surfaces , 2015, SPIE Optical Engineering + Applications.

[17]  Ling Shao,et al.  RGB-D datasets using microsoft kinect or similar sensors: a survey , 2017, Multimedia Tools and Applications.

[18]  Vitaly Kober,et al.  Face recognition based on matching of local features on 3D dynamic range sequences , 2016, Optical Engineering + Applications.

[19]  Truong Q. Nguyen,et al.  Realistic surface geometry reconstruction using a hand-held RGB-D camera , 2016, Machine Vision and Applications.

[20]  Panos M. Pardalos,et al.  Analysis of Images, Social Networks and Texts , 2014, Communications in Computer and Information Science.

[21]  Min-Hyung Choi,et al.  An Improved ICP Algorithm Based on the Sensor Projection for Automatic 3D Registration , 2004, MICAI.

[22]  Jianru Xue,et al.  Improved ICP Algorithm with Bounded Rotation Angle for 2D Point Set Registration , 2014 .

[23]  Kin Hong Wong,et al.  Calibration of multiple Kinect depth sensors for full surface model reconstruction , 2016, International Workshop on Pattern Recognition.

[24]  Juan R. Terven,et al.  A multiple camera calibration and point cloud fusion tool for Kinect V2 , 2017, Sci. Comput. Program..

[25]  Bahari Belaton,et al.  An Improved Iterative Closest Point Algorithm for Rigid Point Registration , 2014, ICMLC.

[26]  Honghai Liu,et al.  Simultaneous Calibration: A Joint Optimization Approach for Multiple Kinect and External Cameras , 2017, Sensors.

[27]  Vitaly Kober,et al.  Impulsive Noise Removal from Color Images with Morphological Filtering , 2017, AIST.

[28]  Seon-Min Rhee,et al.  Two-pass ICP with Color constraint for noisy RGB-D point cloud registration , 2014, 2014 IEEE International Conference on Consumer Electronics (ICCE).

[29]  Hee-Jun Kang,et al.  Improved ICP control algorithm in robot surgery application , 2010, International Forum on Strategic Technology 2010.

[30]  Gérard G. Medioni,et al.  Object modeling by registration of multiple range images , 1991, Proceedings. 1991 IEEE International Conference on Robotics and Automation.

[31]  Vitaly Kober,et al.  Removal of impulse noise clusters from color images with local order statistics , 2017, Optical Engineering + Applications.

[32]  Jun Xie,et al.  Fine registration of 3D point clouds fusing structural and photometric information using an RGB-D camera , 2015, J. Vis. Commun. Image Represent..

[33]  Vitaly Kober,et al.  Fast perceptual image hash based on cascade algorithm , 2017, Optical Engineering + Applications.

[34]  Alexey Ruchay,et al.  Reliable Recognition of Partially Occluded Objects with Correlation Filters , 2018 .