UCSR: Registration and Fusion of Cross-Source 2D and 3D Sensor Data in Unstructured Environments

Heide, Nina Felicitas; Woock, Philipp; Sauer, Maximilian; Leitritz, Timo; Heizmann, Michael

doi:10.23919/fusion45008.2020.9190307

Cited by 5 publications

(8 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…At first, the data is subdivided according to class labels. Outlier filtering using a k-nearest-neighbor search [14] or voxelization of the subdivided data [15] are optional. Benefits are a lower sensitivity to noise, the downside is information loss.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

X³SEG: Model-Agnostic Explanations For The Semantic Segmentation Of 3D Point Clouds With Prototypes And Criticism

Heide

Müller

Petereit

et al. 2021

2021 IEEE International Conference on Image Processing (ICIP)

Self Cite

View full text Add to dashboard Cite

The proposed X 3 Seg approach generates model-agnostic, example-based explanations for the semantic segmentation of 3D point clouds. It retrieves the most similar 3D point sets (prototypes) as well as the most dissimilar point sets (criticism) to the spatially connected 3D point set which is to be explained. X 3 Seg comprises three methods for a holistic understanding of point-by-point class predictions: encompassing, selective, and predictive X 3 Seg. Prototypes and criticism are identified from a particularly generated prototype database by combining different similarity measures. To the best of our knowledge, X 3 Seg is the first model-agnostic explainable artificial intelligence (XAI) approach providing example-based explanations for the semantic segmentation of 3D data with prototypes and criticism. It is demonstrated on RangeNet53++ [1] predictions for 3D point cloud data from the SemanticKITTI dataset [2].

show abstract

Section: Methodsmentioning

confidence: 99%

“…Seq. 01 and 09 are chosen due to their unstructured character [15], seq. 04 is a mixed street scene.…”

Section: Prototype Database and Similarity Measuresmentioning

confidence: 99%

X³SEG: Model-Agnostic Explanations For The Semantic Segmentation Of 3D Point Clouds With Prototypes And Criticism

Heide

Müller

Petereit

et al. 2021

2021 IEEE International Conference on Image Processing (ICIP)

Self Cite

View full text Add to dashboard Cite

show abstract

“…Raw 3D data provides geometric information in 3D space with the position of each measurement point. Following [8] and [9], we regard point density and geometric structure as most conclusive criteria for the IC (IC r3D ). For point density, the density related to the distance from the sensor origin is chosen as the most promising representation [9].…”

Section: Information Content (Ic) Of Raw and Processed Datamentioning

confidence: 99%

“…Following [8] and [9], we regard point density and geometric structure as most conclusive criteria for the IC (IC r3D ). For point density, the density related to the distance from the sensor origin is chosen as the most promising representation [9]. 3D data is transformed from the Cartesian coordinates into homogenized coordinates φ, r, and z: φ = arcsin (y/ x 2 + y 2 ), r = x 2 + y 2 , and z = z.…”

Section: Information Content (Ic) Of Raw and Processed Datamentioning

confidence: 99%

“…3D data is transformed from the Cartesian coordinates into homogenized coordinates φ, r, and z: φ = arcsin (y/ x 2 + y 2 ), r = x 2 + y 2 , and z = z. This yields a more uniform point distribution for active, rotating sensors [9]. A normalization in r and the binning of the homogenized points by their values of φ illustrates the point distribution, and thus the density.…”

Section: Information Content (Ic) Of Raw and Processed Datamentioning

confidence: 99%

See 1 more Smart Citation

A Step towards Explainable Artificial Neural Networks in Image Processing by Dataset Assessment

Heide,

Albrecht,

Heizmann

2020

Tagungsband "Forum Bildverarbeitung 2020

View full text Add to dashboard Cite

We propose a methodology for generalized exploratory data analysis focusing on artificial neural network (ANN) methods. Our method is denoted IC-ACC due to the combined assessment of information content (IC) and accuracy (ACC) and aims at answering a frequently posed question in ANN research: ”What is good data?” As the dataset has the primary influence on the development of the model, IC-ACC provides a step towards explainable ANN methods in the pre-modeling stage by a better insight in the dataset. With this insight, detrimental data can be eliminated before a negative influence on the ANN performance occurs. IC-ACC constitutes a guideline to generate efficient and accurate data for a specific, data-driven ANN method. Moreover, we show that training an ANN for the semantic segmentation of 3D data from unstructured environments with IC-ACC-assessed and -customized training data contributes to a more efficient training. The IC-ACC method is demonstrated on application examples for the visual perception of robotic platforms.

show abstract

Excavation pits—progress estimation and cause of delay identification

et al. 2023

View full text Add to dashboard Cite

This work presents a method for automated excavation speed and progress estimation. First, a measure for the progress speed of an excavation pit is taken from the literature and evaluated regarding the possibility of automation. For each possible parameter, an automated extraction algorithm is presented. The used system is an autonomous excavator arm of a backhoe loader where the used hardware and software system is described. An experimental evaluation of the presented approach has been done with the autonomous system for a small trench, including multiple digging cycles. The resulting measurements seem to include some systematic errors which could be identified and suitable sanity checks could be implemented, removing the erroneous measurements. The remaining measurements were used to determine the excavation speed of the autonomous excavator arm and compared to the values of experienced and amateur operators. The resulting speed and the relevant parameters are then sent to a web application called “DiBa” to notice a potential delay. If a measurement is smaller or larger than expected a fault identification is done in the DiBa.

show abstract

UCSR: Registration and Fusion of Cross-Source 2D and 3D Sensor Data in Unstructured Environments

Cited by 5 publications

References 20 publications

X³SEG: Model-Agnostic Explanations For The Semantic Segmentation Of 3D Point Clouds With Prototypes And Criticism

X³SEG: Model-Agnostic Explanations For The Semantic Segmentation Of 3D Point Clouds With Prototypes And Criticism

A Step towards Explainable Artificial Neural Networks in Image Processing by Dataset Assessment

Excavation pits—progress estimation and cause of delay identification

Contact Info

Product

Resources

About

UCSR: Registration and Fusion of Cross-Source 2D and 3D Sensor Data in Unstructured Environments

Cited by 5 publications

References 20 publications

X3SEG: Model-Agnostic Explanations For The Semantic Segmentation Of 3D Point Clouds With Prototypes And Criticism

X3SEG: Model-Agnostic Explanations For The Semantic Segmentation Of 3D Point Clouds With Prototypes And Criticism

A Step towards Explainable Artificial Neural Networks in Image Processing by Dataset Assessment

Excavation pits—progress estimation and cause of delay identification

Contact Info

Product

Resources

About

X³SEG: Model-Agnostic Explanations For The Semantic Segmentation Of 3D Point Clouds With Prototypes And Criticism

X³SEG: Model-Agnostic Explanations For The Semantic Segmentation Of 3D Point Clouds With Prototypes And Criticism