Efficiency and accuracy of semi-global matching (SGM) make it outperform many stereo matching algorithms and is widely used under challenging occasions. However, SGM only incorporates information along a scanline in each pass and lacks interaction between scanlines, resulting in streak artifacts in the disparity image. We introduce a local edge-aware filtering method to SGM to enhance the interaction of neighboring scanlines, since streak artifacts can be avoided. We use bilateral weights based on intensity similarity and spatial affinity between pixels to build connections among scanlines. In each pass, we recursively estimate the aggregated cost of SGM and compute the weighted average of aggregated costs for pixels in the orthogonal direction to obtain the output of our method along each scanline. As one-dimensional bilateral filtering is used in our method, the extra computation is linear to image resolution and label space, which is a small fraction of that needed by SGM. We present ablation studies using stereo pairs under both constrained and natural conditions to verify the effectiveness of our method. Extensive experiments on Middlebury and Karlsruhe Institute of Technology and Toyota Technology Institute datasets demonstrate that our method removes all streak artifacts, improves the quality of the disparity image, and outperforms many other non-local cost aggregation approaches.
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
Cited By
You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.
You do not have subscription access to this journal. Figure files are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.
You do not have subscription access to this journal. Article tables are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.
You do not have subscription access to this journal. Equations are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.
Error Rate (/%) in Non-Occluded Region of Initial Disparity Image and Corresponding Rankings for Various Cost Aggregation Methods on Middlebury Dataset [6]
Method
GF
NL
ST
MGM
SGM4
SGM8
CSGM4
CSGM8
Aloe
Baby1
Baby2
Baby3
Bowling1
Bowling2
Cloth1
Cloth2
Cloth3
Cloth4
Flowerpots
Lampshade1
Lampshade2
Rocks1
Rocks2
Wood1
Wood2
Art
Books
Cones
Dolls
Laundry
Moebius
Reindeer
Teddy
Tsukuba
Venus
Average
Table 2.
Quantitative Comparison with State-of-the-art Non-local Cost Aggregation Approaches on Middlebury Dataset [6]a
${{\rm Out}_ -}{\rm a}$, percentage of erroneous pixels in all regions; ${{\rm Out}_ -}{\rm n}$, percentage of erroneous pixels in non-occluded regions.
Table 3.
Quantitative Comparison with State-of-the-art Cost Aggregation
Approaches on KITTI 2012 Dataset [41]a
${{\rm
Avg}_{{\rm -a}}}$, average disparity error
in all regions (px); ${{\rm Avg}_
{{\rm -a}}}$, average disparity error
in non-occluded regions (px).
Table 4.
Quantitative Comparison with State-of-the-art Cost Aggregation
Approaches on KITTI 2015 Dataset [45]
“R” and “D” indicate image resolution and disparity space, respectively.
Tables (5)
Table 1.
Error Rate (/%) in Non-Occluded Region of Initial Disparity Image and Corresponding Rankings for Various Cost Aggregation Methods on Middlebury Dataset [6]
Method
GF
NL
ST
MGM
SGM4
SGM8
CSGM4
CSGM8
Aloe
Baby1
Baby2
Baby3
Bowling1
Bowling2
Cloth1
Cloth2
Cloth3
Cloth4
Flowerpots
Lampshade1
Lampshade2
Rocks1
Rocks2
Wood1
Wood2
Art
Books
Cones
Dolls
Laundry
Moebius
Reindeer
Teddy
Tsukuba
Venus
Average
Table 2.
Quantitative Comparison with State-of-the-art Non-local Cost Aggregation Approaches on Middlebury Dataset [6]a
${{\rm Out}_ -}{\rm a}$, percentage of erroneous pixels in all regions; ${{\rm Out}_ -}{\rm n}$, percentage of erroneous pixels in non-occluded regions.
Table 3.
Quantitative Comparison with State-of-the-art Cost Aggregation
Approaches on KITTI 2012 Dataset [41]a
${{\rm
Avg}_{{\rm -a}}}$, average disparity error
in all regions (px); ${{\rm Avg}_
{{\rm -a}}}$, average disparity error
in non-occluded regions (px).
Table 4.
Quantitative Comparison with State-of-the-art Cost Aggregation
Approaches on KITTI 2015 Dataset [45]