Content-Based Guided Image Filtering, Weighted Semi-Global Optimization, and Efficient Disparity Refinement for Fast and Accurate Disparity Estimation

Kordelas, Georgios; Alexiadis, Dimitrios S.; Daras, Petros; Izquierdo, Ebroul

doi:10.1109/tmm.2015.2505905

Cited by 31 publications

(10 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Rating datasets are generally proposed to use in recommendation methods such as collaborative filtering [20], content-based filtering [16], knowledge-based recommendations [4], and Hybrid Recommendations [6]. Moreover, in [26,29], the authors report that the data analyst can also utilize rating datasets by such an appropriate business reason, further they report that rating datasets have privacy violation concerns must be addressed.…”

Section: Rating Datasets [24 26 29]mentioning

confidence: 99%

Achieving Privacy Preservation Constraints in Missing-Value Datasets

2020

View full text Add to dashboard Cite

Privacy violation issues must be taken into consideration when datasets are released for public use. To address these issues, there are various anonymization models to be proposed, e.g., k-anonymity, l-diversity, and t-closeness. However, these anonymization models generally propose to address privacy violation issues in datasets which are assumed that all attributes of them must be completed. Thus, these anonymization models could be insufficient to address privacy violation issues in such a dataset which is allowed to collect missing-values, e.g., rating datasets and trajectory datasets. Therefore, a new appropriate privacy preservation model for missing-value datasets is proposed by this work. With the proposed model, aside from privacy preservation, the data utility is also maintained as much as possible. Moreover, a suitable data utility metric for missing-value datasets is also presented by this work. Furthermore, the proposed model is shown that it is an NP-Complete problem by reduction from the X3C problem.

show abstract

Section: Rating Datasets [24 26 29]mentioning

confidence: 99%

Achieving Privacy Preservation Constraints in Missing-Value Datasets

2020

View full text Add to dashboard Cite

show abstract

“…Again, there possibly be inaccurate disparity generated when utilizing the constant penalties especially at depth discontinuities and occluded regions. Several semi-global matching methods have been proposed to adjust the penalties based on the intensity gradient between neighboring two pixels [22], [35]. However, the pixel with intensity changes may not correspond to disparity discontinuity, and at the same time, the disparity is easy to be seriously interfered by image noise.…”

Section: B Cost Computation and Aggregation With Adaptive Penaltiesmentioning

confidence: 99%

An Efficient Dense Stereo Matching Method for Planetary Rover

Chen

2019

IEEE Access

View full text Add to dashboard Cite

Stereo matching is one of the most important and challenging subjects in the field of planetary rover with a stereo vision system. The generated disparity map can make rover to avoid the obstacle and explore the planetary surface automatically. In this paper, we propose an efficient dense stereo matching method to generate disparity maps for a planetary rover, which relies on 3-D plane fitting, adaptive penalties, and coarse-to-fine disparity constraint. In order to achieve efficient stereo matching at the coarsest level of the pyramid, we present a 3-D plane fitting to reduce the disparity search range and propose adaptive penalties in the more-global matching method for obtaining an accurate disparity map. At the finer level, our method then estimates the disparity search range based on coarse-to-fine disparity constraint and utilizes adaptive penalties to obtain an accurate disparity map. The extensive experiments with stereo images of Chang'e-3 rover demonstrate that our approach can generate disparity maps efficiently and accurately compared with the most state-of-the-art semi-global matching methods, especially at low texture regions, depth discontinuities, and occlusion regions. INDEX TERMS Stereo matching, disparity map, coarse-to-fine architecture, 3-D plane, planetary rover.

show abstract

“…This step can be improved using semi-global or global optimization algorithms such as graph cuts [13], belief propagation [14], or dynamic programming [15,16]. Disparity refinement is done using the same approach as in Mattoccia et al [17] and Kordelas et al [18] to detect occlusions and depth borders. In this step, three consecutive processes are applied: invalid disparity detection, fill-in of invalid disparity values, and median filtering.…”

Section: Related Workmentioning

confidence: 99%

“…Inconsistent pixels between the two disparity maps are marked as having invalid disparities. In our work, we use the same approach defined by Mattoccia et al [17] and Kordelas et al [18]. Disparity values are marked as invalid if they do not satisfy the condition below:…”

Section: Disparity Refinementmentioning

confidence: 99%

Real-time stereo matching on CUDA using Fourier descriptors and dynamic programming

Hammami

Smach²,

Atri

2019

Comp. Visual Media

View full text Add to dashboard Cite

Computation of stereoscopic depth and disparity map extraction are dynamic research topics. A large variety of algorithms has been developed, among which we cite feature matching, moment extraction, and image representation using descriptors to determine a disparity map. This paper proposes a new method for stereo matching based on Fourier descriptors. The robustness of these descriptors under photometric and geometric transformations provides a better representation of a template or a local region in the image. In our work, we specifically use generalized Fourier descriptors to compute a robust cost function. Then, a box filter is applied for cost aggregation to enforce a smoothness constraint between neighboring pixels. Optimization and disparity calculation are done using dynamic programming, with a cost based on similarity between generalized Fourier descriptors using Euclidean distance. This local cost function is used to optimize correspondences. Our stereo matching algorithm is evaluated using the Middlebury stereo benchmark; our approach has been implemented on parallel high-performance graphics hardware using CUDA to accelerate our algorithm, giving a real-time implementation.

show abstract

Content-Based Guided Image Filtering, Weighted Semi-Global Optimization, and Efficient Disparity Refinement for Fast and Accurate Disparity Estimation

Cited by 31 publications

References 49 publications

Achieving Privacy Preservation Constraints in Missing-Value Datasets

Achieving Privacy Preservation Constraints in Missing-Value Datasets

An Efficient Dense Stereo Matching Method for Planetary Rover

Real-time stereo matching on CUDA using Fourier descriptors and dynamic programming

Contact Info

Product

Resources

About