Search Results for author: Marion Neumeier

Found 5 papers, 1 papers with code

Prediction and Interpretation of Vehicle Trajectories in the Graph Spectral Domain

no code implementations16 Aug 2023 Marion Neumeier, Sebastian Dorn, Michael Botsch, Wolfgang Utschick

This work provides a comprehensive analysis and interpretation of the graph spectral representation of traffic scenarios.

Optimization and Interpretability of Graph Attention Networks for Small Sparse Graph Structures in Automotive Applications

1 code implementation25 May 2023 Marion Neumeier, Andreas Tollkühn, Sebastian Dorn, Michael Botsch, Wolfgang Utschick

For automotive applications, the Graph Attention Network (GAT) is a prominently used architecture to include relational information of a traffic scenario during feature embedding.

Graph Attention

A Multidimensional Graph Fourier Transformation Neural Network for Vehicle Trajectory Prediction

no code implementations12 May 2023 Marion Neumeier, Andreas Tollkühn, Michael Botsch, Wolfgang Utschick

This work introduces the multidimensional Graph Fourier Transformation Neural Network (GFTNN) for long-term trajectory predictions on highways.

Descriptive Trajectory Prediction

Gradient Derivation for Learnable Parameters in Graph Attention Networks

no code implementations21 Apr 2023 Marion Neumeier, Andreas Tollkühn, Sebastian Dorn, Michael Botsch, Wolfgang Utschick

This work provides a comprehensive derivation of the parameter gradients for GATv2 [4], a widely used implementation of Graph Attention Networks (GATs).

Graph Attention

Variational Autoencoder-Based Vehicle Trajectory Prediction with an Interpretable Latent Space

no code implementations25 Mar 2021 Marion Neumeier, Andreas Tollkühn, Thomas Berberich, Michael Botsch

This paper introduces the Descriptive Variational Autoencoder (DVAE), an unsupervised and end-to-end trainable neural network for predicting vehicle trajectories that provides partial interpretability.

Decision Making Descriptive +1

Cannot find the paper you are looking for? You can Submit a new open access paper.