Method

sqrtSGM : Memory-Efficient Parametric Semi-Global Matching [sqrtSGM]
[Anonymous Submission]

Submitted on 14 Jun. 2017 08:48 by
[Anonymous Submission]

Running time:7.77 s
Environment:4 cores @ 3.5 Ghz (C/C++)

Method Description:
Parametric aggregation based on GMM which is the
variant of original SGM algorithm. The contribution
of sqrtSGM is that the memory requirement is
extremely lower than that of original SGM without
the accuracy loss. Matching metric is AD-census 9x7
with simple 9x9 box aggregation. And sqrtSGM is
applied for 8 path semi-global aggregation. Finally,
post processing is 3x3 simple median filter.
Parameters:
Num Of GMMmodes : 3
Num Of Samples : 1
threshold Mahalanobis : 2
base learning rate : 0.05
slant surface param. : 0.1
Latex Bibtex:

Detailed Results

This page provides detailed results for the method(s) selected. For the first 20 test images, the percentage of erroneous pixels is depicted in the table. We use the error metric described in Object Scene Flow for Autonomous Vehicles (CVPR 2015), which considers a pixel to be correctly estimated if the disparity or flow end-point error is <3px or <5% (for scene flow this criterion needs to be fulfilled for both disparity maps and the flow map). Underneath, the left input image, the estimated results and the error maps are shown (for disp_0/disp_1/flow/scene_flow, respectively). The error map uses the log-color scale described in Object Scene Flow for Autonomous Vehicles (CVPR 2015), depicting correct estimates (<3px or <5% error) in blue and wrong estimates in red color tones. Dark regions in the error images denote the occluded pixels which fall outside the image boundaries. The false color maps of the results are scaled to the largest ground truth disparity values / flow magnitudes.

Test Set Average

Error D1-bg D1-fg D1-all
All / All 5.53 11.31 6.49
All / Est 5.53 11.31 6.49
Noc / All 4.89 9.89 5.71
Noc / Est 4.89 9.89 5.71
This table as LaTeX

Test Image 0

Error D1-bg D1-fg D1-all
All / All 5.48 4.87 5.40
All / Est 5.48 4.87 5.40
Noc / All 5.23 4.87 5.18
Noc / Est 5.23 4.87 5.18
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 1

Error D1-bg D1-fg D1-all
All / All 3.90 12.24 4.83
All / Est 3.90 12.24 4.83
Noc / All 3.45 12.24 4.45
Noc / Est 3.45 12.24 4.45
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 2

Error D1-bg D1-fg D1-all
All / All 5.46 7.94 5.58
All / Est 5.46 7.94 5.58
Noc / All 4.72 7.94 4.88
Noc / Est 4.72 7.94 4.88
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 3

Error D1-bg D1-fg D1-all
All / All 8.27 14.57 8.86
All / Est 8.27 14.57 8.86
Noc / All 7.60 14.57 8.25
Noc / Est 7.60 14.57 8.25
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 4

Error D1-bg D1-fg D1-all
All / All 8.39 6.60 8.09
All / Est 8.39 6.60 8.09
Noc / All 7.24 6.60 7.14
Noc / Est 7.24 6.60 7.14
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 5

Error D1-bg D1-fg D1-all
All / All 11.53 5.14 10.96
All / Est 11.53 5.14 10.96
Noc / All 9.78 5.14 9.35
Noc / Est 9.78 5.14 9.35
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 6

Error D1-bg D1-fg D1-all
All / All 12.33 2.52 11.30
All / Est 12.33 2.52 11.30
Noc / All 11.59 2.52 10.61
Noc / Est 11.59 2.52 10.61
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 7

Error D1-bg D1-fg D1-all
All / All 3.88 7.02 4.49
All / Est 3.88 7.02 4.49
Noc / All 3.85 7.02 4.48
Noc / Est 3.85 7.02 4.48
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 8

Error D1-bg D1-fg D1-all
All / All 3.74 2.93 3.59
All / Est 3.74 2.93 3.59
Noc / All 3.74 2.93 3.59
Noc / Est 3.74 2.93 3.59
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 9

Error D1-bg D1-fg D1-all
All / All 4.25 5.49 4.57
All / Est 4.25 5.49 4.57
Noc / All 4.26 3.33 4.03
Noc / Est 4.26 3.33 4.03
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 10

Error D1-bg D1-fg D1-all
All / All 4.10 4.15 4.11
All / Est 4.10 4.15 4.11
Noc / All 4.05 4.15 4.07
Noc / Est 4.05 4.15 4.07
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 11

Error D1-bg D1-fg D1-all
All / All 2.66 1.24 2.40
All / Est 2.66 1.24 2.40
Noc / All 2.61 1.24 2.36
Noc / Est 2.61 1.24 2.36
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 12

Error D1-bg D1-fg D1-all
All / All 1.64 2.33 1.69
All / Est 1.64 2.33 1.69
Noc / All 1.38 2.33 1.44
Noc / Est 1.38 2.33 1.44
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 13

Error D1-bg D1-fg D1-all
All / All 1.72 0.93 1.62
All / Est 1.72 0.93 1.62
Noc / All 1.30 0.93 1.25
Noc / Est 1.30 0.93 1.25
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 14

Error D1-bg D1-fg D1-all
All / All 2.34 3.37 2.36
All / Est 2.34 3.37 2.36
Noc / All 1.98 3.37 2.00
Noc / Est 1.98 3.37 2.00
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 15

Error D1-bg D1-fg D1-all
All / All 5.13 3.49 4.98
All / Est 5.13 3.49 4.98
Noc / All 4.70 3.49 4.59
Noc / Est 4.70 3.49 4.59
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 16

Error D1-bg D1-fg D1-all
All / All 6.99 2.02 6.26
All / Est 6.99 2.02 6.26
Noc / All 6.42 2.02 5.77
Noc / Est 6.42 2.02 5.77
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 17

Error D1-bg D1-fg D1-all
All / All 2.57 1.29 2.44
All / Est 2.57 1.29 2.44
Noc / All 1.81 1.29 1.75
Noc / Est 1.81 1.29 1.75
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 18

Error D1-bg D1-fg D1-all
All / All 8.96 8.66 8.82
All / Est 8.96 8.66 8.82
Noc / All 8.39 8.66 8.52
Noc / Est 8.39 8.66 8.52
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 19

Error D1-bg D1-fg D1-all
All / All 2.11 3.91 2.31
All / Est 2.11 3.91 2.31
Noc / All 1.90 3.91 2.13
Noc / Est 1.90 3.91 2.13
This table as LaTeX

Input Image

D1 Result

D1 Error




eXTReMe Tracker