Ensemble Classifier for Combining Stereo Matching Algorithms

Spyropoulos, Aristotle; Mordohai, Philippos

doi:10.1109/3dv.2015.16

Cited by 17 publications

(10 citation statements)

References 42 publications

(49 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…SGM and adaptive local matching) and a dedicated matching function for images with radiometric changes was exploited. A further improvement would be to train an algorithm to choose, or combine, results from several matching algorithms (Spyropoulos and Mordohai, 2015). The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XLI-B5, 2016XXIII ISPRS Congress, 12-19 July 2016 …”

Section: Discussionmentioning

confidence: 99%

A Holistic Approach for Inspection of Civil Infrastructures Based on Computer Vision Techniques

Stentoumis

Protopapadakis

Doulamis

2016

Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

ABSTRACT:In this work, it is examined the 2D recognition and 3D modelling of concrete tunnel cracks, through visual cues. At the time being, the structural integrity inspection of large-scale infrastructures is mainly performed through visual observations by human inspectors, who identify structural defects, rate them and, then, categorize their severity. The described approach targets at minimum human intervention, for autonomous inspection of civil infrastructures. The shortfalls of existing approaches in crack assessment are being addressed by proposing a novel detection scheme. Although efforts have been made in the field, synergies among proposed techniques are still missing. The holistic approach of this paper exploits the state of the art techniques of pattern recognition and stereo-matching, in order to build accurate 3D crack models. The innovation lies in the hybrid approach for the CNN detector initialization, and the use of the modified census transformation for stereo matching along with a binary fusion of two state-of-the-art optimization schemes. The described approach manages to deal with images of harsh radiometry, along with severe radiometric differences in the stereo pair. The effectiveness of this workflow is evaluated on a real dataset gathered in highway and railway tunnels. What is promising is that the computer vision workflow described in this work can be transferred, with adaptations of course, to other infrastructure such as pipelines, bridges and large industrial facilities that are in the need of continuous state assessment during their operational life cycle.

show abstract

Section: Discussionmentioning

confidence: 99%

A Holistic Approach for Inspection of Civil Infrastructures Based on Computer Vision Techniques

Stentoumis

Protopapadakis

Doulamis

2016

Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

show abstract

“…Other authors followed this path addressing efficiency [4,13]. In [22] a CNN is deployed to combine multiple out-of-the-box stereo matchers to obtain more accurate results, inspired by the work of Spyropolous and Mordohai [32] which carried it out by using a random forest. Conversely, Mayer et al [16] proposed DispNet, the first deep architecture for end-to-end stereo computation, completely departing from conventional stereo methodologies.…”

Section: Related Workmentioning

confidence: 99%

Leveraging Confident Points for Accurate Depth Refinement on Embedded Systems

Tosi

Poggi

Mattoccia

2019

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)

View full text Add to dashboard Cite

Despite the notable progress in stereo disparity estimation, algorithms are still prone to errors in challenging conditions. Thus, heuristic disparity refinement techniques are usually deployed to improve accuracy. Moreover, stateof-the-art methods rely on complex CNNs requiring power hungry GPUs not suited for many practical applications constrained by limited computing resources. In this paper, we propose a novel technique for disparity refinement leveraging on confidence measures and a novel, automatic learning-based selection method to discard outliers. Then, a non-local strategy infers missing disparities by analyzing the closest reliable points. This framework is very fast and does not require any hand-tuned thresholding. We assess the performance of our Non-Local Anchoring (NLA), standalone refinement techniques and methods leveraging on confidence measures inside the stereo algorithm. Our evaluation with two popular stereo algorithms shows that our proposal significantly ameliorates their accuracy on Middlebury v3 and KITTI 2015 datasets. Moreover, since our method relies only on cues computed in the disparity domain, it is suited even for COTS stereo cameras coupled with embedded systems, e.g. nVidia Jetson TX2.

show abstract

“…This allows to filter out local outliers from disparity maps and thus to subsequently adjust the ratio of density and reliability. The proposed applications are diverse: Confidence maps are used as weighting-schemes to combine multiple stereo matching algorithms [1,2], different cost functions [3,4,5] or to fuse cost volumes [6] for multi-view stereo, in a reasonable way. Confidence maps are furthermore used to improve the process of depth reconstruction itself: They allow to modulate cost functions in order to adjust the influence of a specific disparity assignment on its neighbours during optimisation [7,8].…”

Section: Introductionmentioning

confidence: 99%

CNN-Based Cost Volume Analysis as Confidence Measure for Dense Matching

Mehltretter

Heipke

2019

2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW)

View full text Add to dashboard Cite

Due to its capability to identify erroneous disparity assignments in dense stereo matching, confidence estimation is beneficial for a wide range of applications, e.g. autonomous driving, which needs a certain degree of confidence as mandatory prerequisite. Especially, the introduction of deep learning based methods resulted in an increasing popularity of this field in recent years, caused by a significantly improved accuracy. Despite this remarkable development, most of these methods rely on features learned from disparity maps only, not taking into account the corresponding 3-dimensional cost volumes. However, it was already demonstrated that with conventional methods based on hand-crafted features this additional information can be used to further increase the accuracy. In order to combine the advantages of deep learning and cost volume based features, in this paper, we propose a novel Convolutional Neural Network (CNN) architecture to directly learn features for confidence estimation from volumetric 3D data. An extensive evaluation on three datasets using three common dense stereo matching techniques demonstrates the generality and state-of-the-art accuracy of the proposed method.

show abstract

Ensemble Classifier for Combining Stereo Matching Algorithms

Cited by 17 publications

References 42 publications

A Holistic Approach for Inspection of Civil Infrastructures Based on Computer Vision Techniques

A Holistic Approach for Inspection of Civil Infrastructures Based on Computer Vision Techniques

Leveraging Confident Points for Accurate Depth Refinement on Embedded Systems

CNN-Based Cost Volume Analysis as Confidence Measure for Dense Matching

Contact Info

Product

Resources

About