Introduction Positioning 3D reconstruction Conclusions Fast field survey with a smartphone A. Masiero F. Fissore, F. Pirotti, A. Guarnieri, A. Vettore CIRGEO – Interdept. Research Center of Geomatics University of Padova – Italy cirgeo@unipd.it 1 Masiero – Fast field survey with a smartphone – 19 Feb 2016, Trieste, Italy
Introduction Positioning 3D reconstruction Conclusions Introduction Mobile Mapping with Smartphones Use of embedded sensors : ● Camera is used as imaging sensor 3D reconstruction of the observed environment via photogrammetry (e.g. SfM) ● Device position estimated by integrating information provided by the embedded sensors spatial referring - Low cost, fast w.r. to other techniques (e.g. TLS) - Limited resources : stringent restrictions on the computational power, limited battery life... ● Goal: exploit information provided by the navigation system to improve the reconstruction procedure 2 Masiero – Fast field survey with a smartphone – 19 Feb 2016, Trieste, Italy 2
Introduction Positioning 3D reconstruction Conclusions Positioning Navigation system Positioning achieved by integrating information: ● GNSS ● inertial sensors (embedded in the device, they provide good local estimates of position variations but drift in long time intervals if used alone) ● WiFi signal strength ● Barometer ● Geometry of the environment Nonlinear filtering 3 Masiero – Fast field survey with a smartphone – 19 Feb 2016, Trieste, Italy 3
Introduction Positioning 3D reconstruction Conclusions Positioning Particle filtering Information fusion of PDR (Pedestrian Dead Reckoning), WiFi, building map... Particle filtering ● Device position is expressed as average position of N particles v u q t + i = q t + s t [ cos α t ] sin α t ● Dynamic equation of each particle: it exploits measured step length and heading direction 4 Masiero – Fast field survey with a smartphone – 19 Feb 2016, Trieste, Italy 4
Introduction Positioning 3D reconstruction Conclusions Positioning Particle filtering ● Advantage: simple to introduce non-linear constraints (and to deal with multiple hypothesis) in position estimation Neglegted, and resampled ● High accuracy for large N, but computational burden issues! ● [Masiero 2014] proposed a revised version of [Widyawan 2012] in order to increase accuracy for small N (N≈100) and uncalibrated sensors ● For further accuracy improvement: - good sensor calibration - exploiting landmarks 5 Masiero – Fast field survey with a smartphone – 19 Feb 2016, Trieste, Italy 5
Introduction Positioning 3D reconstruction Conclusions Positioning Particle filtering ● Information fusion of PDR (Pedestrian Dead Reckoning), WiFi, building map... Particle filtering - [Widyawan 2012]: Particle filter for PDR - [Masiero 2014]: revised version of the particle filter in [Widyawan 2012] in order to increase accuracy for small N (number of particles N≈100) and uncalibrated sensors ● Magnetometer & accelerometer simultaneous calibration [Masiero MMT2015] ● Barometer altitude variation - linear model to describe the relation between pressure and altitude variations ( precision ≈ 0.2m ). 6 Masiero – Fast field survey with a smartphone – 19 Feb 2016, Trieste, Italy 6
Introduction Positioning 3D reconstruction Conclusions 3D reconstruction 3D photogrammetric reconstruction Reconstruction outline ● Compute feature locations (e.g. Harris feature detector) ● Compute feature descriptors (e.g. SIFT) ● Feature matching (Best Bin First – Kd tree search) ● Remove outliers (epipolar geometry, RANSAC or its variants) ● Bundle adjustment (optimize parameter values) Projective reconstruction Control points are used to obtain Euclidean reconstruction and for georeferencing 7 Masiero – Fast field survey with a smartphone – 19 Feb 2016, Trieste, Italy 7
Introduction Positioning 3D reconstruction Conclusions 3D reconstruction 3D photogrammetric reconstruction Reconstruction outline ● Compute feature locations (e.g. Harris feature detector) take into account of ● Compute feature descriptors affine transformations ● Feature matching (Best Bin First – Kd tree search) ● Remove outliers (epipolar geometry, RANSAC or its variants) ● Bundle adjustment (optimize parameter values) Projective reconstruction Control points are used to obtain Euclidean reconstruction and for georeferencing 8 Masiero – Fast field survey with a smartphone – 19 Feb 2016, Trieste, Italy 8
Introduction Positioning 3D reconstruction Conclusions 3D reconstruction Feature matching ● Typically done by using SIFT (Scale-invariant feature transform, [Lowe 1999]) matchings [Vedaldi 2008] ● SIFT deals well with rotations with respect to rotations along the optical axis 9 Masiero – Fast field survey with a smartphone – 19 Feb 2016, Trieste, Italy 9
Introduction Positioning 3D reconstruction Conclusions 3D reconstruction Feature matching ● However, issues can occur when considering other rotations (as typical with generic changes of the point of view) 10 Masiero – Fast field survey with a smartphone – 19 Feb 2016, Trieste, Italy 10
Introduction Positioning 3D reconstruction Conclusions 3D reconstruction Feature matching ● ASIFT [Morel 2011] increases SIFT robustness with respect to such rotations by modelling their effect by means of affine transformations. ● However, in ASIFT 32 affine transformations of each feature are computed 32 2 ≈1000 comparisons between each couple of features in two different images. ● Goal: reducing computational complexity of ASIFT while ensuring increase of matchings with respect to SIFT in the critical cases (e.g. previously described changes of the point of view...) 11 Masiero – Fast field survey with a smartphone – 19 Feb 2016, Trieste, Italy 11
Introduction Positioning 3D reconstruction Conclusions 3D reconstruction Feature matching ● Appearance of a feature seen by camera j depends on the point of view and on the “spatial orientation” of the feature ● Information by the navigation system change of the point of view transformation (translation + rotation) approximately known ● Uncertainty in the “spatial orientation” of the feature 12 Masiero – Fast field survey with a smartphone – 19 Feb 2016, Trieste, Italy 12
Introduction Positioning 3D reconstruction Conclusions 3D reconstruction Feature matching Surface of the real object Image plane Image plane 13 Masiero – Fast field survey with a smartphone – 19 Feb 2016, Trieste, Italy 13
Introduction Positioning 3D reconstruction Conclusions 3D reconstruction Feature matching ● Appearance of a feature seen by camera j depends on the point of view and on the “spatial orientation” of the feature ● Information by the navigation system change of the point of view transformation (translation + rotation) approximately known ● Uncertainty in the “spatial orientation” of the feature ● Compensate for this uncertainty by simulating the effect of 20 possible orientations (on a semi-sphere...) ● Thanks to information provided by the navigation system: - ASIFT: 32 2 ≈1000 comparisons (per feature couple) - Our approach: 20 comparisons (per feature couple) 14 Masiero – Fast field survey with a smartphone – 19 Feb 2016, Trieste, Italy 14
Introduction Positioning 3D reconstruction Conclusions 3D reconstruction Matches with SIFT Images of this example available from the internet [Lhuillier and Quan, 2005] 15 Masiero – Fast field survey with a smartphone – 19 Feb 2016, Trieste, Italy 15
Introduction Positioning 3D reconstruction Conclusions 3D reconstruction Matches with the proposed method Images of this example available from the internet [Lhuillier and Quan, 2005] 16 Masiero – Fast field survey with a smartphone – 19 Feb 2016, Trieste, Italy 16
Introduction Positioning 3D reconstruction Conclusions 3D reconstruction Number of correct matches vs (difference of) observation angle SIFT: Blue x-marks Our approach: red circles 17 Masiero – Fast field survey with a smartphone – 19 Feb 2016, Trieste, Italy 17
Introduction Positioning 3D reconstruction Conclusions 3D reconstruction 3D photogrammetric reconstruction Reconstruction outline ● Compute feature locations (e.g. Harris feature detector) take into account of ● Compute feature descriptors affine transformations ● Feature matching Use approximate epipolar constraints to discard false matchings ● Remove outliers (epipolar geometry, RANSAC or its variants) ● Bundle adjustment (optimize parameter values) Projective reconstruction Control points are used to obtain Euclidean reconstruction and for georeferencing 18 Masiero – Fast field survey with a smartphone – 19 Feb 2016, Trieste, Italy 18
Recommend
More recommend
