74 citations · 360 across the 51 of their papers we have counts for
33 papers · 1 filter
On the Influence of Shape, Texture and Color for Learning Semantic Segmentation
Annika Mütze, Natalie Grabowsky, Edgar Heinert +2
Recent research has investigated the shape and texture biases of pre-trained deep neural networks (DNNs) in image classification. Those works test how much a trained DNN relies on…
Uncertainty and Prediction Quality Estimation for Semantic Segmentation via Graph Neural Networks
Edgar Heinert, Stephan Tilgner, Timo Palm +1
When employing deep neural networks (DNNs) for semantic segmentation in safety-critical applications like automotive perception or medical imaging, it is important to estimate thei…
Reducing Texture Bias of Deep Neural Networks via Edge Enhancing Diffusion
Edgar Heinert, Matthias Rottmann, Kira Maag +1
Convolutional neural networks (CNNs) for image processing tend to focus on localized texture patterns, commonly referred to as texture bias. While most of the previous works in the…
Deep Active Learning with Noisy Oracle in Object Detection
Marius Schubert, Tobias Riedlinger, Karsten Kahl +1
Obtaining annotations for complex computer vision tasks such as object detection is an expensive and time-intense endeavor involving a large number of human workers or expert opini…
LMD: Light-weight Prediction Quality Estimation for Object Detection in Lidar Point Clouds
Tobias Riedlinger, Marius Schubert, Sarina Penquitt +7
Object detection on Lidar point cloud data is a promising technology for autonomous driving and robotics which has seen a significant rise in performance and accuracy during recent…
Identifying Label Errors in Object Detection Datasets by Loss Inspection
Marius Schubert, Tobias Riedlinger, Karsten Kahl +4
Labeling datasets for supervised object detection is a dull and time-consuming task. Errors can be easily introduced during annotation and overlooked during review, yielding inaccu…
Towards Rapid Prototyping and Comparability in Active Learning for Deep Object Detection
Tobias Riedlinger, Marius Schubert, Karsten Kahl +2
Active learning as a paradigm in deep learning is especially important in applications involving intricate perception tasks such as object detection where labels are difficult and…
MGiaD: Multigrid in all dimensions. Efficiency and robustness by coarsening in resolution and channel dimensions
Antonia van Betteray, Matthias Rottmann, Karsten Kahl
Current state-of-the-art deep neural networks for image classification are made up of 10 - 100 million learnable weights and are therefore inherently prone to overfitting. The comp…
Semi-supervised domain adaptation with CycleGAN guided by a downstream task loss
Annika Mütze, Matthias Rottmann, Hanno Gottschalk
Domain adaptation is of huge interest as labeling is an expensive and error-prone task, especially when labels are needed on pixel-level like in semantic segmentation. Therefore, o…
Automated Detection of Label Errors in Semantic Segmentation Datasets via Deep Learning and Uncertainty Quantification
Matthias Rottmann, Marco Reese
In this work, we for the first time present a method for detecting label errors in image datasets with semantic segmentation, i.e., pixel-wise class labels. Annotation acquisition…
False Negative Reduction in Semantic Segmentation under Domain Shift using Depth Estimation
Kira Maag, Matthias Rottmann
State-of-the-art deep neural networks demonstrate outstanding performance in semantic segmentation. However, their performance is tied to the domain represented by the training dat…
Uncertainty Quantification and Resource-Demanding Computer Vision Applications of Deep Learning
Julian Burghoff, Robin Chan, Hanno Gottschalk +4
Bringing deep neural networks (DNNs) into safety critical applications such as automated driving, medical imaging and finance, requires a thorough treatment of the model's uncertai…
Detecting and Learning the Unknown in Semantic Segmentation
Robin Chan, Svenja Uhlemeyer, Matthias Rottmann +1
Semantic segmentation is a crucial component for perception in automated driving. Deep neural networks (DNNs) are commonly used for this task and they are usually trained on a clos…
UQGAN: A Unified Model for Uncertainty Quantification of Deep Classifiers trained via Conditional GANs
Philipp Oberdiek, Gernot A. Fink, Matthias Rottmann
We present an approach to quantifying both aleatoric and epistemic uncertainty for deep neural networks in image classification, based on generative adversarial networks (GANs). Wh…
Towards Unsupervised Open World Semantic Segmentation
Svenja Uhlemeyer, Matthias Rottmann, Hanno Gottschalk
For the semantic segmentation of images, state-of-the-art deep neural networks (DNNs) achieve high segmentation accuracy if that task is restricted to a closed set of classes. Howe…
False Positive Detection and Prediction Quality Estimation for LiDAR Point Cloud Segmentation
Pascal Colling, Matthias Rottmann, Lutz Roese-Koerner +1
We present a novel post-processing tool for semantic segmentation of LiDAR point cloud data, called LidarMetaSeg, which estimates the prediction quality segmentwise. For this purpo…
Background-Foreground Segmentation for Interior Sensing in Automotive Industry
Claudia Drygala, Matthias Rottmann, Hanno Gottschalk +2
To ensure safety in automated driving, the correct perception of the situation inside the car is as important as its environment. Thus, seat occupancy detection and classification…
Gradient-Based Quantification of Epistemic Uncertainty for Deep Object Detectors
Tobias Riedlinger, Matthias Rottmann, Marius Schubert +1
The vast majority of uncertainty quantification methods for deep object detectors such as variational inference are based on the network output. Here, we study gradient-based epist…
Validation of Simulation-Based Testing: Bypassing Domain Shift with Label-to-Image Synthesis
Julia Rosenzweig, Eduardo Brito, Hans-Ulrich Kobialka +8
Many machine learning applications can benefit from simulated data for systematic validation - in particular if real-life data is difficult to obtain or annotate. However, since si…
SegmentMeIfYouCan: A Benchmark for Anomaly Segmentation
Robin Chan, Krzysztof Lis, Svenja Uhlemeyer +6
State-of-the-art semantic or instance segmentation deep neural networks (DNNs) are usually trained on a closed set of semantic classes. As such, they are ill-equipped to handle pre…
Improving Video Instance Segmentation by Light-weight Temporal Uncertainty Estimates
Kira Maag, Matthias Rottmann, Serin Varghese +3
Instance segmentation with neural networks is an essential task in environment perception. In many works, it has been observed that neural networks can predict false positive insta…
Entropy Maximization and Meta Classification for Out-Of-Distribution Detection in Semantic Segmentation
Robin Chan, Matthias Rottmann, Hanno Gottschalk
Deep neural networks (DNNs) for the semantic segmentation of images are usually trained to operate on a predefined closed set of object classes. This is in contrast to the "open wo…
YOdar: Uncertainty-based Sensor Fusion for Vehicle Detection with Camera and Radar Sensors
Kamil Kowol, Matthias Rottmann, Stefan Bracke +1
In this work, we present an uncertainty-based method for sensor fusion with camera and radar data. The outputs of two neural networks, one processing camera and the other one radar…
MetaDetect: Uncertainty Quantification and Prediction Quality Estimates for Object Detection
Marius Schubert, Karsten Kahl, Matthias Rottmann
In object detection with deep neural networks, the box-wise objectness score tends to be overconfident, sometimes even indicating high confidence in presence of inaccurate predicti…