Method

Semantic-Geometric Duality Framework [SGDuality]
[Anonymous Submission]

Submitted on 17 Jul. 2026 04:18 by
[Anonymous Submission]

Running time:0.02 s
Environment:GPU @ 1.5 Ghz (Python)

Method Description:
This method jointly performs semantic segmentation
and stereo disparity estimation. Geometry-guided
Semantic Modulation injects disparity-derived
geometric cues into the semantic branch, while
Semantic-guided Disparity Refinement uses semantic
cues to selectively refine disparity estimates. A
Boundary Geometry Coupling loss provides class-
pair-aware supervision at semantic boundaries
during training without increasing inference cost.
The full model contains 8.22M parameters and runs
at 67.37 FPS on an NVIDIA A100 GPU.
Parameters:
max_disp=192; GSM window K=5; alpha, beta, and
gamma are learned;
lambda_seg=lambda_disp=lambda_bgc=1; lambda_b=1;
lambda_g=0.5; tau=2% of the training-set semantic
boundary samples.
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 1.55 3.03 1.80
All / Est 1.55 3.03 1.80
Noc / All 1.45 2.84 1.68
Noc / Est 1.45 2.84 1.68
This table as LaTeX

Test Image 0

Error D1-bg D1-fg D1-all
All / All 1.48 1.49 1.48
All / Est 1.48 1.49 1.48
Noc / All 1.48 1.49 1.48
Noc / Est 1.48 1.49 1.48
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 1

Error D1-bg D1-fg D1-all
All / All 1.58 2.95 1.73
All / Est 1.58 2.95 1.73
Noc / All 1.51 2.95 1.67
Noc / Est 1.51 2.95 1.67
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 2

Error D1-bg D1-fg D1-all
All / All 1.99 6.69 2.22
All / Est 1.99 6.69 2.22
Noc / All 1.93 6.69 2.17
Noc / Est 1.93 6.69 2.17
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 3

Error D1-bg D1-fg D1-all
All / All 2.35 1.80 2.30
All / Est 2.35 1.80 2.30
Noc / All 2.33 1.80 2.28
Noc / Est 2.33 1.80 2.28
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 4

Error D1-bg D1-fg D1-all
All / All 1.23 3.11 1.54
All / Est 1.23 3.11 1.54
Noc / All 1.20 3.11 1.53
Noc / Est 1.20 3.11 1.53
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 5

Error D1-bg D1-fg D1-all
All / All 2.05 1.87 2.03
All / Est 2.05 1.87 2.03
Noc / All 1.96 1.87 1.95
Noc / Est 1.96 1.87 1.95
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 6

Error D1-bg D1-fg D1-all
All / All 2.23 0.49 2.05
All / Est 2.23 0.49 2.05
Noc / All 2.28 0.49 2.09
Noc / Est 2.28 0.49 2.09
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 7

Error D1-bg D1-fg D1-all
All / All 0.38 2.70 0.83
All / Est 0.38 2.70 0.83
Noc / All 0.38 2.70 0.84
Noc / Est 0.38 2.70 0.84
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 8

Error D1-bg D1-fg D1-all
All / All 0.36 2.07 0.68
All / Est 0.36 2.07 0.68
Noc / All 0.36 2.07 0.67
Noc / Est 0.36 2.07 0.67
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 9

Error D1-bg D1-fg D1-all
All / All 0.43 2.41 0.94
All / Est 0.43 2.41 0.94
Noc / All 0.43 2.53 0.95
Noc / Est 0.43 2.53 0.95
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 10

Error D1-bg D1-fg D1-all
All / All 1.12 1.45 1.20
All / Est 1.12 1.45 1.20
Noc / All 1.12 1.45 1.20
Noc / Est 1.12 1.45 1.20
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 11

Error D1-bg D1-fg D1-all
All / All 1.19 0.28 1.03
All / Est 1.19 0.28 1.03
Noc / All 1.20 0.28 1.03
Noc / Est 1.20 0.28 1.03
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 12

Error D1-bg D1-fg D1-all
All / All 0.83 0.64 0.82
All / Est 0.83 0.64 0.82
Noc / All 0.70 0.64 0.69
Noc / Est 0.70 0.64 0.69
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 13

Error D1-bg D1-fg D1-all
All / All 0.36 0.11 0.33
All / Est 0.36 0.11 0.33
Noc / All 0.33 0.11 0.30
Noc / Est 0.33 0.11 0.30
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 14

Error D1-bg D1-fg D1-all
All / All 1.37 0.00 1.35
All / Est 1.37 0.00 1.35
Noc / All 1.25 0.00 1.23
Noc / Est 1.25 0.00 1.23
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 15

Error D1-bg D1-fg D1-all
All / All 2.38 0.34 2.20
All / Est 2.38 0.34 2.20
Noc / All 2.44 0.34 2.24
Noc / Est 2.44 0.34 2.24
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 16

Error D1-bg D1-fg D1-all
All / All 3.15 0.09 2.70
All / Est 3.15 0.09 2.70
Noc / All 2.97 0.09 2.54
Noc / Est 2.97 0.09 2.54
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 17

Error D1-bg D1-fg D1-all
All / All 0.83 0.26 0.77
All / Est 0.83 0.26 0.77
Noc / All 0.82 0.26 0.76
Noc / Est 0.82 0.26 0.76
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 18

Error D1-bg D1-fg D1-all
All / All 5.23 1.01 3.22
All / Est 5.23 1.01 3.22
Noc / All 5.15 1.01 3.17
Noc / Est 5.15 1.01 3.17
This table as LaTeX

Input Image

D1 Result

D1 Error


Test Image 19

Error D1-bg D1-fg D1-all
All / All 0.76 0.16 0.69
All / Est 0.76 0.16 0.69
Noc / All 0.75 0.16 0.68
Noc / Est 0.75 0.16 0.68
This table as LaTeX

Input Image

D1 Result

D1 Error




eXTReMe Tracker