F
Fabian Timm
Researcher at Bosch
Publications - 35
Citations - 1336
Fabian Timm is an academic researcher from Bosch. The author has contributed to research in topics: Object detection & Probabilistic logic. The author has an hindex of 9, co-authored 35 publications receiving 769 citations. Previous affiliations of Fabian Timm include University of Lübeck.
Papers
More filters
Journal ArticleDOI
Deep Multi-Modal Object Detection and Semantic Segmentation for Autonomous Driving: Datasets, Methods, and Challenges
Di Feng,Christian Haase-Schutz,Lars Rosenbaum,Heinz Hertlein,Claudius Gläser,Fabian Timm,Werner Wiesbeck,Klaus Dietmayer +7 more
TL;DR: In this article, the authors systematically summarize methodologies and discuss challenges for deep multi-modal object detection and semantic segmentation in autonomous driving and provide an overview of on-board sensors on test vehicles, open datasets, and background information for object detection.
Proceedings Article
Accurate eye centre localisation by means of gradients
Fabian Timm,Erhardt Barth +1 more
TL;DR: This work proposes an approach for accurate and robust eye centre localisation by using image gradients using a simple objective function, which only consists of dot products, and demonstrates that this method yields a significant improvement regarding both accuracy and robustness.
Proceedings ArticleDOI
Leveraging Heteroscedastic Aleatoric Uncertainties for Robust Real-Time LiDAR 3D Object Detection
TL;DR: In this article, a multi-loss function is designed to incorporate uncertainty estimations predicted by auxiliary output layers to improve the accuracy of LiDAR 3D object detection. But, the proposed method ignores to train from noisy samples, and focuses more on informative ones.
Proceedings ArticleDOI
Non-parametric texture defect detection using Weibull features
Fabian Timm,Erhardt Barth +1 more
TL;DR: A novel, non-parametric approach for defect detection in textures that only employs two features is proposed, which can detect local deviations of texture images in an unsupervised manner with high accuracy and can be applied for real-time applications.
Posted Content
Can We Trust You? On Calibration of a Probabilistic Object Detector for Autonomous Driving.
TL;DR: This work identifies uncertainty miscalibration problems in a probabilistic LiDAR 3D object detection network, and proposes three practical methods to significantly reduce errors in uncertainty calibration.