Matching cost function using robust soft rank transformations

Dinh, Vinh Quang; Pham, Cuong Cao; Jeon, Jae Wook

doi:10.1049/iet-ipr.2015.0736

Cited by 9 publications

(6 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Various stereo correspondence algorithms have been used in various fields such as medical, satellite-based earth observation, space exploration, autonomous robots, and security systems. However, solving the stereo correspondence problem is still a challenging task in certain areas due to texture-less regions [4], occlusions, illumination variations, changes int the weather such as snow, sun, and rain [3], and cameras with different focal lengths [5].…”

Section: B Stereo Matchingmentioning

confidence: 99%

Multitasking Correlation Network for Depth Information Reconstruction

Nguyen

Hoang

Hong

2023

REV J. Electron. Commun.

View full text Add to dashboard Cite

In this paper, we propose a novel multi-tasking network for stereo matching. The proposed network is trained to approximate similarity functions in statistics and linear algebra such as correlation coefficient, distance correlation and cosine similarity. By doing this, the proposed method decreases the amount of time needed to calculate the disparity map by using CNN's ability to calculate multiple pairs of image patches at the same time. We then compare the execution time and overall accuracy between the traditional method using functions and our method. The results show the model's ability to mimic the traditional method's performance while taking considerably less time to perform the task.

show abstract

Section: B Stereo Matchingmentioning

confidence: 99%

Multitasking Correlation Network for Depth Information Reconstruction

Nguyen

Hoang

Hong

2023

REV J. Electron. Commun.

View full text Add to dashboard Cite

show abstract

“…Matching cost functions that exploit ordinal values rather can tolerate this kind of intensity transformation. These matching cost functions include the rank and census transforms [25], the support local binary pattern (SLBP) [26], the fuzzy encoding pattern [27], and the soft rank transform [28].…”

Section: Related Workmentioning

confidence: 99%

Unsupervised Learning for Stereo Matching Using Single-View Videos

Hong

Ahn

2020

IEEE Access

View full text Add to dashboard Cite

This paper proposes an unsupervised approach to construct a deep learning based stereo matching method using single-view videos (SMV). From videos, a set of corresponding points are computed between images, and image patches that center at the computed points are extracted. Negative and positive samples constitute a dataset to train a similarity network that is then used as a matching cost function. In addition, we propose a local-global matching cost network that exploits the first feature maps (local features) accompanying with last feature maps (global features) as output feature of the proposed network. The concatenated features are connected to full-connected layers and the network outputs a similarity measure of an image patch pair as a matching cost. Computed matching costs are aggregated using semiglobal matching and cross-based cost aggregation, followed by sub-pixel interpolation, left-right consistency check, median and bilateral filtering. We evaluate the proposed stereo matching methods using popular stereo matching datasets, including KITTI 2012 and 2015, and Middlebury. We submit the disparity maps to their benchmark servers to evaluate the performance of SMV. We also compared the generalization of SMV and baseline methods using the training sets of the three datasets. The benchmark results show that SMV is the most accurate method among unsupervised approach, and it even outperforms several deep learning based stereo matching using supervised manner. The evaluation results of generalization show that SMV is comparative with the baseline method, MC-CNN, which is trained with supervision. INDEX TERMS Stereo matching, unsupervised learning, video extraction.

show abstract

“…The last one is based on the similarity of pixel features in two camera images of binocular vision. Representative algorithms of this type of method include the CENSUS algorithm [18], SLBP algorithm (Support Local binary pattern) [19], FEP algorithm (Fuzzy encoding pattern) [20], and RSRT algorithm (Robust soft binary transformation) [21]. FEP was the best one of these algorithms.…”

Section: Introductionmentioning

confidence: 99%

Binocular visual dimension measurement method for rectangular workpiece with a precise stereoscopic matching algorithm

Chen¹,

Zhong²,

Ze³

et al. 2022

Meas. Sci. Technol.

View full text Add to dashboard Cite

Binocular vision measurement benefits from high prediction robustness and low structural complexity. However, there are still significant flaws in its accuracy. In this paper, binocular vision measurement of a rectangular workpiece is investigated. A new precise measurement method based on binocular vision is designed to achieve precise measurement of rectangular workpiece dimensions. Firstly, an algorithm for workpiece location based on Zernike moments and corner matching is proposed and employed to precisely locate the workpiece and extract the sub-pixel coordinates of discrete points on an image's edge. Then, a novel stereoscopic matching algorithm combined with epipolar-geometry and cross-ratio invariance (CMEC) is proposed to improve the accuracy of binocular vision stereoscopic matching. Finally, a projection plane is introduced after the 3D reconstruction of discrete points in the workpiece contours by fitting the plane with least squares. The projection plane limits the coordinate fluctuations of discrete points. Furthermore, the data screening is used to further improve the accuracy of size calculation. The experimental results of the standard checkerboard and actual workpiece show that CMEC's matching accuracy reached 99%, and the proposed method's measurement accuracy reached 0.018 mm. This work presents a novel algorithm for stereoscopic matching in binocular vision and machine vision measurement.

show abstract

Matching cost function using robust soft rank transformations

Cited by 9 publications

References 23 publications

Multitasking Correlation Network for Depth Information Reconstruction

Multitasking Correlation Network for Depth Information Reconstruction

Unsupervised Learning for Stereo Matching Using Single-View Videos

Binocular visual dimension measurement method for rectangular workpiece with a precise stereoscopic matching algorithm

Contact Info

Product

Resources

About