object indentification using 3d sketchup models in environment scans
DESCRIPTION
Abstract. This work introduces a novel solution for localizing objects based on search strings and freely available Google SketchUp models. To this end we automatically download and preprocess a collection of 3D models to obtain equivalent point clouds. The outdoor scan is segmented into individual objects, which are sequentially matched with the models by a variant of iterative closest points algorithm using seven degrees of freedom and resulting in a highly precise pose estimation of the object. An error function evaluates the similarity level. The approach is verified using various segmented cars and their corresponding 3D models.TRANSCRIPT
Object Identification Using 3D SketchUp Models in
Environment Scans
Flavia Grosan, Alexandru TandrauProf. Dr. Andreas Nüchter
Thursday, May 12, 2011
Introductionn “I can’t find my Audi A4. Bot, please find it for me!”
Thursday, May 12, 2011
Introduction
n SLAM
n Laser range scanners
n ICP
n Semantics
Thursday, May 12, 2011
State of the Art
n Horn - closed form solution for ICP
n translation, rotation, scale
n Nüchter - semantic mapping
n determine coarse features: walls, floors
n trained classifier to identify more delicate objects
Thursday, May 12, 2011
State of the Artn Object Localization
n Li-Jia - 2D object localization
n Meger - Semantic Robot Challenge
n Kestler - probabilistic representation
n manual labeling needed
n maintains internal neural net trained data
n Lai and Fox - Google Warehouse to train classifiers
n Albrecht - CAD models and ontologiesThursday, May 12, 2011
Scientific Contributionn Combine laser scanning with object detection and
localization
n Simple scan matching instead of classifiers and probabilistic approaches
n Evaluates Google 3D Warehouse - a new, large 3D model database
Thursday, May 12, 2011
From Model to Point Cloud
n Google 3D Warehouse - collection of user made SketchUp models
n A model is composed of:
n Faces
n ComponentInstances
n Groups
Thursday, May 12, 2011
n Additional sampling procedure needed
n Add random points inside each triangular face proportionally to its area
n Center the point cloud around its centroid and bound the coordinates in [-α, α]
From Model to Point Cloud
Thursday, May 12, 2011
Model - Scan Matching
Is the model present in the scan? If so, where?Thursday, May 12, 2011
Model - Scan Matchingn Ground Removal
n Stiene et al.
n Compute gradient between closest points in the same vertical sweep plane
n A point is classified as ground if -θ ≤ αi,j ≤ θ
Thursday, May 12, 2011
Model - Scan Matching
Object segmentation by region growingwith starting point
Thursday, May 12, 2011
Model - Scan Matching
Object segmentation by region growingwith starting point
Thursday, May 12, 2011
n The centroid of the segmented object is the new system origin. The object coordinates are bounded in [-α, α] (center and scale step)
n Modified ICP (SICP) to match model and scan
n Scale
n Favor rotations on the y-axis (wheels on the ground)
n Points are linked in both directions (scan to model, model to scan)
n Recover transformation matrix to original scan
Model - Scan Matching
Thursday, May 12, 2011
Model - Scan Matching
SICP animationThursday, May 12, 2011
Model - Scan Matching
SICP animationThursday, May 12, 2011
Model - Scan Matching
n S - the scan points, M - the model points
n c(p) ∈ M - the model point which is closest to p ∈ S
the error function penalizes points in the scanwhich have no correspondence in the model
Thursday, May 12, 2011
Experiments and Resultsn acquired scans using a Riegl VZ-400 3D laser
scanner in the Jacobs University parking lot
n segmented 5 different cars based on starting points
n automatically downloaded relevant Google SketchUp models and pre-processed them (resampling, scale & center)
n SICP with 4 starting rotations around the vertical axis
Thursday, May 12, 2011
Mercedes C350
n 8920 points in scan
n 89 models available in Google Warehouse
Thursday, May 12, 2011
Mercedes C350 - best models
Thursday, May 12, 2011
Mercedes C350 - worst models
Thursday, May 12, 2011
Audi A4
n 18801 points in scan
n 80 models available in Google Warehouse
Thursday, May 12, 2011
Audi A4 - best models
Thursday, May 12, 2011
Audi A4 - worst models
Thursday, May 12, 2011
Audi A4 in entire scan
Thursday, May 12, 2011
Mercedes C350 vs different brand models
Thursday, May 12, 2011
Conclusionsn Segmented objects above 10,000 points behaved well in SICP
n controlled in practice by taking more scans
n Number of Google Warehouse models
n Volkswagen Golf - 200+ models
n Citroen C5 - 11 models
n ~ 80 models needed to get good matches
0 10 20 30 40
Audi A4
VW Golf
Mercedes C350
Perfect Match %
Thursday, May 12, 2011
Conclusionsn Identified Volkswagen Golf - an older variant
Older Golf63%
Newer Golf37%
Best matches distribution VW Golf
n SICP ranks higher Google models resembling the old Golf version
n SICP identifies similar shapes
n Mercedes C350 vs. other car brands
Thursday, May 12, 2011
Conclusions
n SICP solves the goal finding problem
n Automatic scan segmentation
n Correct identification of Audi A4
n Next 2 matches - also cars, different brands
Thursday, May 12, 2011
Future Work
n Refine model search in Google Warehouse
n Improve the error function
n Tackle indoor scenarios
n SICP - extendable to full-scene understanding
n Create an online platform for SICP
n Integrate SICP as a plugin for RoboEarth
Thursday, May 12, 2011