Fast template matching and pose estimation in 3D point clouds

Vock, Richard; Dieckmann, Alexander; Ochmann, Sebastian; Klein, Reinhard

doi:10.1016/j.cag.2018.12.007

Cited by 53 publications

(35 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…, and the registration errors in the previous iteration) are used for registration and assignment. The method is attractive since it reduces the point-cloud-to-model assignment into a templatematching problem [48,49] and aims to also minimize possible target-based registration errors. The method, however, does not incorporate the impact of instrumental measurement errors pertaining to the point's spatial uncertainties (see Section 2).…”

Section: Point Cloud Analysis: Case Of Negligible Construction Errorsmentioning

confidence: 99%

Field Information Modeling (FIM)™: Best Practices Using Point Clouds

Maalek

2021

Remote Sensing

View full text Add to dashboard Cite

This study presented established methods, along with new algorithmic developments, to automate point cloud processing in support of the Field Information Modeling (FIM)™ framework. More specifically, given a multi-dimensional (n-D) designed information model, and the point cloud’s spatial uncertainty, the problem of automatic assignment of point clouds to their corresponding model elements was considered. The methods addressed two classes of field conditions, namely (i) negligible construction errors and (ii) the existence of construction errors. Emphasis was given to defining the assumptions, potentials, and limitations of each method in practical settings. Considering the shortcomings of current frameworks, three generic algorithms were designed to address the point-cloud-to-model assignment. The algorithms include new developments for (i) point cloud vs. model comparison (negligible construction errors), (ii) robust point neighborhood definition, and (iii) Monte-Carlo-based point-cloud-to-model surface hypothesis testing (existence of construction errors). The effectiveness of the new methods was demonstrated in real-world point clouds, acquired from construction projects, with promising results. For the overall problem of point-cloud-to-model assignment, the proposed point cloud vs. model and point-cloud-to-model hypothesis testing methods achieved F-measures of 99.3% and 98.4%, respectively, on real-world datasets.

show abstract

Section: Point Cloud Analysis: Case Of Negligible Construction Errorsmentioning

confidence: 99%

Field Information Modeling (FIM)™: Best Practices Using Point Clouds

Maalek

2021

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…, and the registration errors in the previous iteration-are used for registration and assignment. The method is attractive since it reduces the point cloud to model assignment into a template matching problem [38,39], and aims to also minimize possible target-based registration errors. The method, however, does not incorporate the impact of instrumental measurement errors pertaining to the point's spatial uncertainties (see Section 2).…”

Section: Point Cloud Analysis: Case Of Negligible Construction Errorsmentioning

confidence: 99%

“…Fitting the model directly to the point cloud, which aims at finding the group of points that match the geometry of the element. This can be accomplished by means of heuristic methods such as template matching [38,39], and robust least squares adjustment [27,45,46]. 2.…”

Section: Point Cloud Analysis: Existence Of Construction Errorsmentioning

confidence: 99%

“…Once the points are designated (classified) to the desired surface model (e.g. plane) using Algorithm 2, Algorithm 1 (or template matching [38]) can be utilized on only the designated points for final point to model assignment. The final consideration for Algorithm 2 is to define a reliable neighborhood around each point, and to determine the required number of Monte Carlo iterations, , to form a reliable distribution of RMSE, which are discussed in the following.…”

Section: Local Point Cloud To Model Hypothesis Testingmentioning

confidence: 99%

See 1 more Smart Citation

Field Information Modeling (FIM)™: Best Practices using Point Clouds

Maalek¹

2021

Preprint

View full text Add to dashboard Cite

This study presents established methods, along with new algorithmic developments, to automate the point cloud processing within the Field Information Modeling (FIM)™ framework. More specifically, given an n-D designed information model, and the point cloud’s spatial uncertainty, the problem of automatic assignment of the point clouds to their corresponding elements within the designed model is considered. The methods addressed two classes of field conditions, namely, (i) negligible construction errors; and (ii) existence of construction errors. The emphasis was given to describing and defining the assumptions in each method and shed light on some of their potentials and limitations in practical settings. Considering the shortcomings of current point cloud processing frameworks, three new and generic algorithms were developed to help solve the point cloud to model assignment in field conditions with both negligible, and existence (or speculation) of construction errors. The effectiveness of the new methods was demonstrated in real-world point clouds, acquired from construction projects, with promising results.

show abstract

“…Targets Recognition service is based on the 3D Registration template matching approach, which compares the received input frame against a set of predefined images, called models. In [10] is specified that the template matching approach uses information extracted from real object images to identify the corresponding pattern. Moreover, the authors highlight that this method can recognize partial occurrences.…”

Section: Architecturementioning

confidence: 99%

The Prototype of a Traffic Signs Recognition System Using Mobile Augmented Reality

Diță¹,

Enăchescu²

2019

Proceedings of the 18th International Conference on INFORMATICS in ECONOMY Education, Research and Business Technologies

View full text Add to dashboard Cite

In this paper is proposed a traffic signs recognition prototype based on template matching approach of the 3D Registration technique in order to make use of the benefits provided by mobile augmented reality. The proposed solution finds speed traffic signs, extracted from a mobile device camera used in real environment, and uses them as AR markers to augment the virtual elements. This information is meant to help the driver during driving process by giving him various audio or visual indications via his mobile device. As one of the top three causes of road accidents is the distracted driving, usually caused by the driver talking on the mobile phone or sending a text message, including the audio component that gives signals when the road signs are not observed is useful. We proposed an ontology to attach some knowledge to each traffic sign, to guide its interpretation. Nevertheless, the performance of the prototype is affected by several factors, such as: video source, lighting conditions, occlusion, chaotic background and viewing angle, deterioration and deformations of the traffic signs etc. In order to reduce the negative effects of above factors is necessary to use the following methods: image resizing, grayscale transformation, median blur, Hough transformation. In the first part of the research we focused on describing the architecture of the prototype and the components of the OpenCV library, which are integrated in the current solution. In the second phase, we highlighted the most common issues that affect the performance of the application.

show abstract

Fast template matching and pose estimation in 3D point clouds

Cited by 53 publications

References 9 publications

Field Information Modeling (FIM)™: Best Practices Using Point Clouds

Field Information Modeling (FIM)™: Best Practices Using Point Clouds

Field Information Modeling (FIM)™: Best Practices using Point Clouds

The Prototype of a Traffic Signs Recognition System Using Mobile Augmented Reality

Contact Info

Product

Resources

About