On the Impact of the Error Measure Selection in Evaluating Disparity
MapsIvan Cabezas, Victor Padilla, Maria Trujillo and
Margaret [email protected]
June 27th 2012World Automation Congress, ISIAC, Puerto Vallarta - Mexico
Slide 2
Multimedia and Vision Laboratory
MMV is a research group of the Universidad del Valle in Cali, Colombia
Ivan MariaVictor Margaret
On the Impact of the Error Measure Selection in Evaluating Disparity maps, WAC – ISIAC, 2012, Puerto Vallarta, MexicoMultimedia and Vision Laboratory Research: http://mmv-lab.univalle.edu.co
CameraSystem
3D World
2D Images
InverseProblem
OpticsProblem
Slide 3
Content
Stereo Vision Application Domains The Impact of Inaccurate Disparity Estimation Quantitative Evaluation Commonly Used Evaluation Measures Error Measure Function Error Measures Purpose and Meaning Research Problem Comparative Performance Scenario
Middlebury's Evaluation Model A* Evaluation Model Research Questions Algorithm to Measure the Consistency Consistency According to Evaluation Models
Conclusions
On the Impact of the Error Measure Selection in Evaluating Disparity maps, WAC – ISIAC, 2012, Puerto Vallarta, Mexico
Slide 4
Stereo Vision
The stereo vision problem is to recover the 3D structure of a scene
3D ModelStereo Images
Disparity Map
Left Right
Correspondence Algorithm
ReconstructionAlgorithm
Yang Q. et al., Stereo Matching with Colour-Weighted Correlation, Hierarchical Belief Propagation, and Occlusion Handling, IEEE PAMI 2009Scharstein D., and Szeliski R., High-accuracy Stereo Depth Maps using Structured Light, CVPR 2003On the Impact of the Error Measure Selection in Evaluating Disparity maps, WAC – ISIAC, 2012, Puerto Vallarta, Mexico
Points Disparity Values
d: P L 01
dmax
.
.
.
23
p1p2
.
.
.
p3
pn
P L
P
CrCl
πl πrpl pr
Bf
Z
Slide 5
Applications Domains
3D recovering has multiple application domains
On the Impact of the Error Measure Selection in Evaluating Disparity maps, WAC – ISIAC, 2012, Puerto Vallarta, Mexico
Whitehorn M., Vincent T., Debrunner C. and Steele J., Stereo Vision on LHD Automation References, IEEE, Trans on Industry Apps., 2003Van der Mark W., and Gavrila D., Real-Time Dense Stereo for Intelligent Vehicles, IEEE Trans. On Intelligent Transportation Systems, 2006Point Grey Research Inc., www.ptgrey.com
Slide 6
The Impact of Inaccurate Disparity Estimation
Disparity is the distance between corresponding points
Trucco, E. and Verri A., Introductory Techniques for 3D Computer Vision, Prentice Hall 1998
Accurate Disparity Estimation Inaccurate Disparity Estimation
pr’
P ’
P
CrCl
πl πrpl pr
B
f
Z
On the Impact of the Error Measure Selection in Evaluating Disparity maps, WAC – ISIAC, 2012, Puerto Vallarta, Mexico
P
CrCl
πl πrpl pr
B
f
ZZ ’
Slide 7
Quantitative Evaluation
Szeliski, R., Prediction Error as a Quality Metric for Motion and Stereo, ICCV 2000Kostliva, J., Cech, J., and Sara, R., Feasibility Boundary in Dense and Semi-Dense Stereo Matching, CVPR 2007Tomabari, F., Mattoccia, S., and Di Stefano, L., Stereo for robots: Quantitative Evaluation of Efficient and Low-memory Dense Stereo Algorithms, ICCARV 2010Cabezas, I. and Trujillo M., A Non-Linear Quantitative Evaluation Approach for Disparity Estimation, VISAPP 2011Cabezas, I. Trujillo M., and Florian M., An Evaluation Methodology for Stereo Correspondence Algorithms, VISAPP 2012
The use of a methodology allows to:
Assert specific components and procedures
Tune algorithm's parameters
Measure the progress in the field
On the Impact of the Error Measure Selection in Evaluating Disparity maps, WAC – ISIAC, 2012, Puerto Vallarta, Mexico
Slide 8
Commonly Used Evaluation Measures
There are different evaluation measures
On the Impact of the Error Measure Selection in Evaluating Disparity maps, WAC – ISIAC, 2012, Puerto Vallarta, Mexico
Sigma Z Error, SZE
Cabezas, I., Padilla, V., and Trujillo M., A Measure for Accuracy Disparity Maps Evaluation, CIARP 2011
Slide 9
Error Measure Function
On the Impact of the Error Measure Selection in Evaluating Disparity maps, WAC – ISIAC, 2012, Puerto Vallarta, Mexico
Yang Q. et al., Stereo Matching with Colour-Weighted Correlation, Hierarchical Belief Propagation, and Occlusion Handling, IEEE PAMI 2008Scharstein D., and Szeliski R., High-accuracy Stereo Depth Maps using Structured Light, CVPR 2003
all
nonocc
disc
Measure nonocc all discMAE 0,41 1,48 0,70
MSE 1,48 33,97 4,25
MRE 0,01 0,03 0,02
BMP 2,90 8,78 7,79
SZE 71,39 341,55 37,86
Estimated Ground-truth
Test Bed
Error Criteria
Evaluation Measures
× ×
Slide 10
Error Measures Purpose and Meaning
On the Impact of the Error Measure Selection in Evaluating Disparity maps, WAC – ISIAC, 2012, Puerto Vallarta, Mexico
In practice, different error measures are used for a same purpose: find a distance between estimated and ground-truth disparity data
They have different meaning, as well as different properties
Slide 11
Research Problem
On the Impact of the Error Measure Selection in Evaluating Disparity maps, WAC – ISIAC, 2012, Puerto Vallarta, Mexico
The use of different error measures may produce contradictories score errors
Scharstein, D. and Szeliski, R., A Taxonomy and Evaluation of Dense Two-Frame Stereo Correspondence Algorithms, IJCV 2002Scharstein, D. and Szeliski, R., http://vision.middlebury.edu/stereo/eval/, 2012
RDPADCensus
Algorithms
Teddy
Cones
Slide 12
Comparative Performance Scenario
On the Impact of the Error Measure Selection in Evaluating Disparity maps, WAC – ISIAC, 2012, Puerto Vallarta, Mexico
Four stereo image pairs: Tsukuba, Venus, Teddy, Cones
Three error criteria: nonocc, all, disc
112 Stereo Correspondence Algorithms
Two evaluation models: Middlebury and A* k: a threshold for determining the top-performer
algorithms in the Middlebury's evaluation model
Slide 13
Middlebury’s Methodology Evaluation Model
Compute Error Measures
Algorithm nonocc all discObjectStereo 2.20 1 6.99 2 6.36 1
GC+SegmBorder 4.99 5 5.78 1 8.66 5PUTv3 2.40 2 9.11 5 6.56 2
PatchMatch 2.47 3 7.80 3 7.11 3ImproveSubPix 2.96 4 8.22 4 8.55 4
Algorithm Average Rank
FinalRank
ObjectStereo 1.33 1
PatchMatch 3.00 2PUTv3 3.33 3
GC+SegmBorder 3,66 4ImproveSubPix 4.00 5
Apply Evaluation Model
Algorithm nonocc all discObjectStereo 2.20 6.99 6.36
GC+SegmBorder 4.99 5.78 8.66PUTv3 2.40 9.11 6.56
PatchMatch 2.47 7.80 7.11ImproveSubPix 2.96 8.22 8.55
Scharstein, D. and Szeliski, R., http://vision.middlebury.edu/stereo/eval/, 2012
Middlebury’sEvaluation Model
…
On the Impact of the Error Measure Selection in Evaluating Disparity maps, WAC – ISIAC, 2012, Puerto Vallarta, Mexico
Slide 14
A* Evaluation Model
The A* evaluation model performs a partitioning of the stereo algorithms under evaluation, based on the Pareto Dominance relation
Compute Error Measures
Algorithm nonocc all discObjectStereo 2.20 6.99 6.36
GC+SegmBorder 4.99 5.78 8.66PUTv3 2.40 9.11 6.56
PatchMatch 2.47 7.80 7.11ImproveSubPix 2.96 8.22 8.55
Algorithm nonocc all disc SetObjectStereo 2.20 6.99 6.36 A*
GC+SegmBorder 4.99 5.78 8.66 A*PUTv3 2.40 9.11 6.56 A’
PatchMatch 2.47 7.80 7.11 A’ImproveSubPix 2.96 8.22 8.55 A’
Apply Evaluation Model
, GC+SegmBorder
PatchMatch
ObjectStereo
PUTv3 ImproveSubPix,,
On the Impact of the Error Measure Selection in Evaluating Disparity maps, WAC – ISIAC, 2012, Puerto Vallarta, Mexico
…
A* Evaluation Model
Slide 15
Research Questions
On the Impact of the Error Measure Selection in Evaluating Disparity maps, WAC – ISIAC, 2012, Puerto Vallarta, Mexico
What is the impact of using an error measure instead of other? Different evaluation results are obtained using different evaluation measures
Scharstein, D. and Szeliski, R., A Taxonomy and Evaluation of Dense Two-Frame Stereo Correspondence Algorithms, IJCV 2002Scharstein, D. and Szeliski, R., http://vision.middlebury.edu/stereo/eval/, 2012
Middlebury's Model A* Model
Slide 16
Automati c is computed without human intervention
Reliable I has to operate without being influenced by external factors, and in a
deterministic way
Meaningful is intended for a particular purpose, has a concise interpretation and does not lead to
ambiguous results
Unbiased is capable of accomplish the measurements for which is was conceived, and its use allow to
perform impartial comparisons
Consistent The scores produced by an error measure should be compatible with produced scores by another error
measure with a common particular purpose
Research Questions (ii)
How does an error measure have to be choose ? A characterisation of error measures may serve as selection criteria An error measure:
AUTOMATIC
RELIABLE
CONSISTENT
MEANINGFUL
UNBIASED
On the Impact of the Error Measure Selection in Evaluating Disparity maps, WAC – ISIAC, 2012, Puerto Vallarta, Mexico
Slide 17
Algorithm to Measure the Consistency
On the Impact of the Error Measure Selection in Evaluating Disparity maps, WAC – ISIAC, 2012, Puerto Vallarta, Mexico
Consistency is measured by determining the percentages of agreements in obtained results by varying the used error measure
Slide 18
Consistency According to Evaluation Models
On the Impact of the Error Measure Selection in Evaluating Disparity maps, WAC – ISIAC, 2012, Puerto Vallarta, Mexico
The MRE, followed by the MSE error measures shown the highest percentages of consistency using the Middlebury's model
The SZE, followed by the MRE error measures shown the highest percentages of consistency using the A* model
Middlebury's Model A* Model
Slide 19
Conclusions
On the Impact of the Error Measure Selection in Evaluating Disparity maps, WAC – ISIAC, 2012, Puerto Vallarta, Mexico
Using the Middlebury’s evaluation model the MRE and the MSE shown a high consistency
Using the A* evaluation model the SZE and the MRE shown a high consistency
The BMP shown a low consistency in both used evaluation models
A characterisation of error measure was presented in order to support the selection of an error measure
It includes the following attributes: automatic, reliable, meaningful, unbiased, and consistent
Experimental evaluation was focused on measuring consistency
The selection of an error measure is not a trivial issue since it impacts on obtained results during a disparity maps evaluation process
On the Impact of the Error Measure Selection in Evaluating Disparity
MapsIvan Cabezas, Victor Padilla, Maria Trujillo and
Margaret [email protected]
June 27th 2012World Automation Congress, ISIAC, Puerto Vallarta - Mexico