Efficient Stereo Matching Based on a New Confidence Metric Won-Hee Lee, Yumi Kim, and Jong Beom Ra Department of Electrical Engineering, KAIST, Daejeon, Korea 20th European Signal Processing Conference (EUSIPCO 2012)
Jan 29, 2016
Efficient Stereo Matching Based on a New Confidence Metric
Won-Hee Lee, Yumi Kim, and Jong Beom Ra
Department of Electrical Engineering, KAIST, Daejeon, Korea
20th European Signal Processing Conference (EUSIPCO 2012)
2
Outline• Introduction
• Related Work
• Proposed Algorithm
• Experimental Results
• Conclusion
3
Introduction
4
Introduction• For the TV application, stereo matching should be
performed in real-time.
• Aggregation kernel size is to be small• Aggregation process takes large computation loads • May cause problems in a textureless area
• Texture area information incorrect textureless area
• Propose a new confidence metric for stereo matching• For efficient refinement (with small kernel size)
Objective:
5
Introduction
35X35
5X35
[4] K. Zhang, J. Lu, and G. Lafruit, “Cross-based local stereo matching using orthogonal integral images,” IEEE TCSVT, 2009.
6
Related Work
7
Related Work• Cross-based stereo matching algorithm[4]
• Raw matching cost:
• Aggregated cost:
Ud(x) : local support region : the number of pixels in Ud(x)
8
Related Work• Cross-based stereo matching algorithm[4]
• Winner-take-all:
d0(x) : the initial disparitydmax(x) : the maximum disparity
9
Related Work• Confidence metrics[5]:
• Several metrics were proposed to measure the confidence level of match
• Utilizing:
• Aggregated cost
• Curvature of the cost curve
• Left-right consistency
[5] X. Hu and P. Mordohai, “Evaluation of stereo confidence indoors and outdoors,” in CVPR, 2010
10
Confidence metrics• 1) Matching score metric (MSM)
C : aggregated costdi : the disparity that reveals the ith minimum cost
White(High confidence)
Black(Low confidence)
11
Confidence metrics• 2) Curvature of cost curve metric (CUR)
12
Confidence metrics• 3) Naive peak ratio metric (PKRN)
13
Confidence metrics• 4) Naive winner margin metric (WMNN)
• computes a margin between two minimum costs • normalize it with the sum of total costs
14
Confidence metrics• 5) Left right difference metric (LRD)
min{cR(x - d1, dR)}: the minimum value of a cost curve at the corresponding pixel in the right image.
15
ProposedAlgorithm
16
Framework
17
Proposed Confidence Metric
‧:Correct estimated pixels X : Incorrect estimated pixels
18
Proposed Confidence Metric• The new metric is proposed as
• Characteristic:• extracts the curvature information across a range larger than that
including three cost values in the CUR metric
• : improve the metric performance for a cost graph with a small curvature.
LoG : a Laplacian of Gaussian filter of n-taps
19
Refinement• Weighted median filter
• Weight :
: the initial disparity of neighboring pixels (same color segment)
: duplication operator
offset a slope of function
20
Refinement• Histogram-based color segmentation algorithm[6]:
[6] J. Delon, A. Desolneux, J. L. Lisani, and A. B. Petro, “A nonparametric approach for histogram segmentation,” IEEE TIP, 2007.
21
Refinement• The filtering is applied only to the limited number if pixels
• Due to small size of filtering kernel
• To enlarge the filtering range
• Vertically propagate the filtered result of a current pixel
A
B
C
(current)
If Weightpropagate > WeightB
DisparityB = Disparitypropagate
Else Datapropagate = DataB
Datapropagate = DataAFiltered Disparity
Weight
Color segment
index
Propagation Data After weighted median filtering…
22
ExperimentalResults
23
Experimental Results• Parameters:
Aggregation kernel
Filtering kernel
T n σ τ
5 x 35 5 x 63 60 7 10 2
n : Laplacian of Gaussian filter of n-taps
offset a slope of function
24
Experimental ResultsBad pixel Confidence mapInitial disparity map
25
Experimental Results AUC: Area Under the Curve
Venus Tsukuba
Teddy Cones
26
Experimental Results
Error rate (Threshold = 1)
Error rate (Threshold = 1)
Cross-based
Adaptivesupport-weight
27
Experimental Results[4]
[10]
Proposed
After Aggregation
After Aggregation
After Aggregation
31
Conclusion
32
Conclusion• Presented an efficient stereo matching algorithm
• Applying a weighted median filter that is based on the proposed confidence metric.
• Successfully refine initial disparities.
• Competitive to the existing algorithms with a large size of aggregation kernel.