1

Foundation Models for Amodal Video Instance Segmentation in Automated Driving

In this work, we study amodal video instance segmentation for automated driving. Previous works perform amodal video instance …

Jasmin Breitenstein, Franz Jünger, Andreas Bär, Tim Fingscheidt

Non-Causal to Causal SSL-Supported Transfer Learning: Towards A High-Performance Low-Latency Speech Vocoder

Recently, BigVGAN has emerged as high-performance speech vocoder. Its sequence-to-sequence-based synthesis, however, prohibits usage in …

Renzheng Shi, Andreas Bär, Marvin Sach, Wouter Tirry, Tim Fingscheidt

Frozen Feature Augmentation for Few-Shot Image Classification

Training a linear classifier or lightweight model on top of pretrained vision model outputs, so-called ‘frozen features’, …

Andreas Bär, Neil Houlsby, Mostafa Dehghani, Manoj Kumar

A Novel Benchmark for Refinement of Noisy Localization Labels in Autolabeled Datasets for Object Detection

Autolabeling approaches are attractive w.r.t. time and cost as they allow fast annotation without human intervention. However, can we …

Andreas Bär, Jonas Uhrig, Jeethesh Pai Umesh, Marius Cordts, Tim Fingscheidt

Improvements to Image Reconstruction-Based Performance Prediction for Semantic Segmentation in Highly Automated Driving

The performance of deep neural networks is typically measured with ground truth data which is expensive and not available during …

Andreas Bär, Daniel Kusuma, Tim Fingscheidt

Detecting Adversarial Perturbations in Multi-Task Perception

While deep neural networks (DNNs) achieve impressive performance on environment perception tasks, their sensitivity to adversarial …

Marvin Klingner, Varun Ravi-Kumar, Senthil Yogamani, Andreas Bär, Tim Fingscheidt

Adaptive Bitrate Quantization Scheme Without Codebook for Learned Image Compression

We propose a generic approach to quantization without codebook in learned image compression called onehot max (OHM, Ω) quantization. It …

Jonas Löhdefink, Jonas Sitzmann, Andreas Bär, Tim Fingscheidt

Performance Prediction for Semantic Segmentation by a Self-Supervised Image Reconstruction Decoder

In supervised learning, a deep neural network’s performance is measured using ground truth data. In semantic segmentation, ground truth …

Andreas Bär, Marvin Klingner, Jonas Löhdefink, Fabian Hüger, Peter Schlicht, Tim Fingscheidt

From a Fourier-Domain Perspective on Adversarial Examples to a Wiener Filter Defense for Semantic Segmentation

Despite recent advancements, deep neural networks are not robust against adversarial perturbations. Many of the proposed adversarial …

Nikhil Kapoor, Andreas Bär, Serin Varghese, Jan David Schneider, Fabian Hüger, Peter Schlicht, Tim Fingscheidt

Detection of Collective Anomalies in Images for Automated Driving Using an Earth Mover’s Deviation (EMDEV) Measure

For visual perception in automated driving, a reliable detection of so-called corner cases is important. Corner cases appear in many …

Jasmin Breitenstien, Andreas Bär, Daniel Lipinski, Tim Fingscheidt