2016
DOI: 10.1109/tcsvt.2015.2473375
|View full text |Cite
|
Sign up to set email alerts
|

Accurate Image-Guided Stereo Matching With Efficient Matching Cost and Disparity Refinement

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1

Citation Types

0
36
0

Year Published

2016
2016
2020
2020

Publication Types

Select...
7
2

Relationship

0
9

Authors

Journals

citations
Cited by 88 publications
(36 citation statements)
references
References 35 publications
0
36
0
Order By: Relevance
“…This method [49] is similar to ours in terms of multi-scale cost-volume utilization, but its purpose is to improve the quality of the disparity maps, not to reduce the computational complexity. Recently, Zhan et al [48] proposed some techniques for local stereo matching methods to improve the accuracy: mask filtering as a pre-processing, an improved matching cost function, and multi-step disparity refinement as a post-processing. Inspired by the great success of convolutional neural networks (CNNs) in image recognition task, CNNs are recently used for computing the label costs (matching costs in stereo matching and optical flow estimation) instead of hand-crafted cost functions [11,13,23,46,47], which has led to significant improvement in terms of accuracy.…”
Section: Cost Aggregation Methods For Labeling Problemsmentioning
confidence: 99%
“…This method [49] is similar to ours in terms of multi-scale cost-volume utilization, but its purpose is to improve the quality of the disparity maps, not to reduce the computational complexity. Recently, Zhan et al [48] proposed some techniques for local stereo matching methods to improve the accuracy: mask filtering as a pre-processing, an improved matching cost function, and multi-step disparity refinement as a post-processing. Inspired by the great success of convolutional neural networks (CNNs) in image recognition task, CNNs are recently used for computing the label costs (matching costs in stereo matching and optical flow estimation) instead of hand-crafted cost functions [11,13,23,46,47], which has led to significant improvement in terms of accuracy.…”
Section: Cost Aggregation Methods For Labeling Problemsmentioning
confidence: 99%
“…The 4 th ranked method KADI [47] presents a two-phase strategy for combining separate cost volumes, a mean-shift segmentation-driven approach for handling disparity outliers and disparity histogram analysis for fostering low-textured area plane fitting. Notice that [50] was under review at the time of this paper's writing. Therefore, the proposed method ranks 4 th among already published methods.…”
Section: B Middlebury Online Stereo Evaluation Benchmarkmentioning
confidence: 99%
“…Usually it is difficult to obtain full-FoV depth by traditional stereo-matching methods [1,2]. This kind of method can be formulated as a three-step pipeline including matching cost calculation [3], cost aggregation/optimization [4,5], and disparity refinement [6,7].…”
Section: Introductionmentioning
confidence: 99%