On accurate dense stereo-matching using a local adaptive multi-cost approach

Stentoumis, Christos; Γραμματικόπουλος, Λάζαρος; Kalisperakis, I.; Karras, George C.

doi:10.1016/j.isprsjprs.2014.02.006

Cited by 61 publications

(51 citation statements)

References 54 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Bilateral filtering can also be applied to the minimum spanning tree, to achieve non-local aggregation and linear time complexity (Yang, 2015). Weighting can also be applied to non-parametric census transform through duplicating code (Spangenberg et al, 2013) or adaptively modifying the aggregation window (Stentoumis et al, 2014). It is possible to preserve geometric discontinuities through global methods by loosening the smoothness constraints through parameter tuning, e.g., P1 and P2 for SGM (Hirschmuller, 2008), and  for graph cut (Kolmogorov and Zabih, 2001).…”

Section: Related Workmentioning

confidence: 99%

“…Using adaptive filtering to preserve sharp features is in general only suitable for local methods, which are usually less accurate than global methods (Stentoumis et al, 2014). However, in global methods it is contradictory to both enforce smoothness and preserve discontinuities in complex scenes by a single parameter set.…”

Section: Related Workmentioning

confidence: 99%

“…Second, a texture-aware SGM is proposed to adaptively tune the parameters, to both enforce smoothness and preserve sharp edges. Because textureless areas not only represent similar intensities but also disparity continuities and vice versa (Stentoumis et al, 2014;Yang, 2015), this study quantitatively measures the texture information then self-tunes the parameters of SGM accordingly.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Texture-Aware Dense Image Matching Using Ternary Census Transform

Huang¹,

Chen²,

Wu³

et al. 2016

ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

ABSTRACT:Textureless and geometric discontinuities are major problems in state-of-the-art dense image matching methods, as they can cause visually significant noise and the loss of sharp features. Binary census transform is one of the best matching cost methods but in textureless areas, where the intensity values are similar, it suffers from small random noises. Global optimization for disparity computation is inherently sensitive to parameter tuning in complex urban scenes, and must compromise between smoothness and discontinuities. The aim of this study is to provide a method to overcome these issues in dense image matching, by extending the industry proven Semi-Global Matching through 1) developing a ternary census transform, which takes three outputs in a single order comparison and encodes the results in two bits rather than one, and also 2) by using texture-information to self-tune the parameters, which both preserves sharp edges and enforces smoothness when necessary. Experimental results using various datasets from different platforms have shown that the visual qualities of the triangulated point clouds in urban areas can be largely improved by these proposed methods. Corresponding Author.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Texture-Aware Dense Image Matching Using Ternary Census Transform

Huang¹,

Chen²,

Wu³

et al. 2016

ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

show abstract

“…Nebiker completed the reconstruction of historical aerial imagery and the subsequent analysis of settlement change [15]. Stentoumis proposed a local stereo method with high computational efficiency and achieved accurate 3D reconstruction under various scenarios [16]. Meng proposed a new matching method based on best seed propagation considering neighborhood disparity constraints [17].…”

Section: Introductionmentioning

confidence: 99%

An Optimized Method for Terrain Reconstruction Based on Descent Images

Zheng

et al. 2016

J. Eng. Technol. Sci.

View full text Add to dashboard Cite

Abstract. An optimization method is proposed to perform high-accuracy terrain reconstruction of the landing area of Chang'e III. First, feature matching is conducted using geometric model constraints. Then, the initial terrain is obtained and the initial normal vector of each point is solved on the basis of the initial terrain. By changing the vector around the initial normal vector in small steps a set of new vectors is obtained. By combining these vectors with the direction of light and camera, the functions are set up on the basis of a surface reflection model. Then, a series of gray values is derived by solving the equations. The new optimized vector is recorded when the obtained gray value is closest to the corresponding pixel. Finally, the optimized terrain is obtained after iteration of the vector field. Experiments were conducted using the laboratory images and descent images of Chang'e III. The results showed that the performance of the proposed method was better than that of the classical feature matching method. It can provide a reference for terrain reconstruction of the landing area in subsequent moon exploration missions. Keywords:Chang'e III; descent images; feature point extraction; geometric constraint; terrain reconstruction. IntroductionIn lunar exploration, high-precision terrain reconstruction of the landing area has significance for achieving the scientific objectives of future projects and is important for safety and efficient path planning of the rover [1]. The Chang'e II detector acquired a large number of lunar images and other information in previous lunar exploration projects. Digital orthophoto maps with 1.5 m resolution and digital elevation models with 4 m resolution are currently being produced using the existing images captured by the Chang'e II satellite [2][3][4]. However, a terrain reconstruction with this accuracy can only be used for macro scientific investigation and cannot be used for path planning of the rover of Chang'e III [5]. The navigation cameras attached to the Chang'e III rover have a high resolution and can provide the basis for high-precision terrain reconstruction. However, the use of navigation cameras was not conducive to

show abstract

“…Stereo matching has always been a focus in the field of stereo vision research [4,5] . The stereo matching algorithms can be categorized into three types: area-based matching algorithm [6,7] , phase-based matching algorithm [8,9] and feature-based matching algorithm [10][11][12][13] .…”

mentioning

confidence: 99%