CrossPatch-Based Rolling Label Expansion for Dense Stereo Matching

Xu, Huaiyuan; Chen, Xiao; Liang, Hongying; Ren, Siyu; Wang, Yi; Cai, Huaiyu

doi:10.1109/access.2020.2985106

Cited by 10 publications

(8 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The drawback is that there are many shared regions for filtering, which would cause huge redundant computations. CRLE [18] adopts a strategy similar to that of Local-Exp, but they use the cross patches. There are many overlapping patches in 𝛼-expansion for LocalExp and CRLE, which result in redundant computations.…”

Section: Related Workmentioning

confidence: 99%

“…"A95" and "A90"are 95%, and 99% error quantile in pixels. In this study, a comparison is drawn on the proposed method and some state-of-the-art disparity refining methods, including LocalExp [17], CRLE [18], 3DMST [27], PMSC [16] and deep learning stereo methods (AANet [30], GANet [32] and MC-CNN-acrt [28]). Due to the fact that the deep learning methods would utilize abundant images for training on training set, the deep learning methods are subject to a comparison with respect to the test set.…”

Section: Evaluation On Middlebury Benchmark V3mentioning

confidence: 99%

“…In the proposed method, the feature point optimization is employed to construct the candidate label set for each pixel quickly, instead of calculating per-pixel label from huge 3D labels space. However, in terms of LocalExp [17], CRLE [18], 3DMST [27] and PMSC [16], each label from the huge space is calculated, and the subsequent optimization is performed, which induced huge redundant computations. As shown in Table 1, the proposed method can reach the lower error rates compared with 3DMST and PMSC under many evaluating metric with respect to the training set and test set, and can also reach the lower error rates compared with deep learning methods with respect to the test set.…”

Section: Evaluation On Middlebury Benchmark V3mentioning

confidence: 99%

See 2 more Smart Citations

LESC: Superpixel cut‐based local expansion for accurate stereo matching

Cheng

Zhao

Yang

et al. 2021

IET Image Processing

View full text Add to dashboard Cite

The rapid estimation of the accurate disparity between pixels is the goal of stereo matching. However, it is very difficult for the 3D labels-based methods due to huge search space of 3D labels, especially for high-resolution images. In this paper, a novel superpixel cut-based method is proposed, in an attempt to get the accurate disparity map efficiently, including the multi-layer superpixel optimization and iteractive local 𝛼-expansion in parallel. As for the multi-layer superpixel optimization, feature point optimization is designed to get accurate candidate labels that are set for most pixels using non-local cost aggregation strategy and update per-pixel labels of the corresponding superpixels from the candidate label sets on the small-size superpixel layer, and then update the middle to large-size superpixel layers progressively using non-local cost aggregation strategy. In order to provide more prior information to identify weak texture and textureless regions in non-local cost aggregation, the weight combination of "intensity + gradient + binary image" is proposed for constructing an optimal minimum spanning tree (MST) to calculate the aggregated matching cost and obtain the labels of minimum aggregated matching cost. Moreover, the local patch surrounding the corresponding superpixels is designed to accelerate superpixel optimization in parallel, and a neighborhood structure is presented to optimize the algorithm in this study, including superpixel neighborhood and patch neighborhood. As for the iteractive local 𝛼-expansion, three layers of patch structure corresponding to the superpixel neighborhood structure is proposed for optimizing the algorithm in this study. The experimental results show that higher accuracy could be achieved via the method in this study compared with some known state-of-the-art stereo methods on KITTI 2015 and Middlebury benchmark V3, which are the standard benchmarks for testing the stereo matching methods.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Evaluation On Middlebury Benchmark V3mentioning

confidence: 99%

Section: Evaluation On Middlebury Benchmark V3mentioning

confidence: 99%

See 1 more Smart Citation

LESC: Superpixel cut‐based local expansion for accurate stereo matching

Cheng

Zhao

Yang

et al. 2021

IET Image Processing

View full text Add to dashboard Cite

show abstract

“…Image matching is a basic task in the computer vision field. Traditional image matching including stereo matching [1][2][3] and optical flow [4,5] establishes a dense correspondence field between two photos in the same scene based on photo-geometric consistency. However, semantic matching is different, as it establishes the correspondence field between two images based on semantic consistency [6][7][8][9][10], in other words, it looks for the point pair with the same semantics across two images.…”

Section: Introductionmentioning

confidence: 99%

Semantic Matching Based on Semantic Segmentation and Neighborhood Consensus

Xu¹,

Chen²,

Cai³

et al. 2021

Applied Sciences

Self Cite

View full text Add to dashboard Cite

Establishing dense correspondences across semantically similar images is a challenging task, due to the large intra-class variation caused by the unconstrained setting of images, which is prone to cause matching errors. To suppress potential matching ambiguity, NCNet explores the neighborhood consensus pattern in the 4D space of all possible correspondences, which is based on the assumption that the correspondence is continuous in space. We retain the neighborhood consensus constraint, while introducing semantic segmentation information into the features, which makes them more distinguishable and reduces matching ambiguity from a feature perspective. Specifically, we combine the semantic segmentation network to extract semantic features and the 4D convolution to explore 4D-space context consistency. Experiments demonstrate that our algorithm has good semantic matching performances and semantic segmentation information can improve semantic matching accuracy.

show abstract

“…With the development of computer science and multimedia technology, 3D video and free viewpoint video (FVV) have drawn more attention. Compared to 2D video, 3D video introduces depth information, which can provide more immersive viewing experience to viewers [1], [2]. As the ultimate of 3D video, FVV allows the viewer to freely choose the viewpoint within a certain range [3].…”

Section: Introductionmentioning

confidence: 99%

Local Foreground Removal Disocclusion Filling Method for View Synthesis

Liang

Chen

et al. 2020

IEEE Access

Self Cite

View full text Add to dashboard Cite

View synthesis is an effective method to generate the contents of multiple views based on a limited number of reference views, which can be used in 2D to 3D conversion, free viewpoint video and multiview video rendering. Depth-image-based rendering (DIBR) is a practical technique to generate virtual view by using a 2D reference view and its depth image. However, a critical problem in DIBR process is that disocclusions might be produced in the synthesized image because the background occluded by the foreground objects in the reference view may be exposed in the virtual view. In this paper, a local foreground removal method is proposed for disocclusion filling. Morphology-based depth image preprocessing is performed before DIBR, aiming to correct the depth value of the ghosts and remove ghost artifacts. In the synthesized virtual image, pixels on the disocclusion edge are identified and classified. Then they are positioned in the reference image by inverse 3D warping. Local foreground regions that occlude the corresponding background are removed from both the reference image and its depth image based on the disocclusion edge pixels. Removed region is filled with surrounding background contents, and depth information is used in this process to prevent foreground penetration. The predicted background contents are warped to the disocclusion region, thereby achieving the hole filling. Experimental results show that the proposed method performs better than the other methods in disocclusion filling, and improves the subjective and objective quality of the synthesized view. In the evaluation results of PSNR, SSIM, FSIMc and VSI, our method improves by 0.

show abstract

CrossPatch-Based Rolling Label Expansion for Dense Stereo Matching

Cited by 10 publications

References 47 publications

LESC: Superpixel cut‐based local expansion for accurate stereo matching

LESC: Superpixel cut‐based local expansion for accurate stereo matching

Semantic Matching Based on Semantic Segmentation and Neighborhood Consensus

Local Foreground Removal Disocclusion Filling Method for View Synthesis

Contact Info

Product

Resources

About