papers

Publications (5)

cs.LG2024

Examining Common Paradigms in Multi-Task Learning

Cathrin Elich, Lukas Kirchdorfer, Jan M. Köhler +1

While multi-task learning (MTL) has gained significant attention in recent years, its underlying mechanisms remain poorly understood. Recent methods did not yield consistent perfor…

cs.CV2022

Weakly Supervised Learning of Multi-Object 3D Scene Decompositions Using Deep Shape Priors

Cathrin Elich, Martin R. Oswald, Marc Pollefeys +1

Representing scenes at the granularity of objects is a prerequisite for scene understanding and decision making. We propose PriSMONet, a novel approach based on Prior Shape knowled…

cs.LG2024

Analytical Uncertainty-Based Loss Weighting in Multi-Task Learning

Lukas Kirchdorfer, Cathrin Elich, Simon Kutsche +3

With the rise of neural networks in various domains, multi-task learning (MTL) gained significant relevance. A key challenge in MTL is balancing individual task losses during neura…

cs.CV2023

Learning-based Relational Object Matching Across Views

Cathrin Elich, Iro Armeni, Martin R. Oswald +2

Intelligent robots require object-level scene understanding to reason about possible tasks and interactions with the environment. Moreover, many perception tasks such as scene reco…

cs.CV2019

3D-BEVIS: Bird's-Eye-View Instance Segmentation

Cathrin Elich, Francis Engelmann, Theodora Kontogianni +1

Recent deep learning models achieve impressive results on 3D scene analysis tasks by operating directly on unstructured point clouds. A lot of progress was made in the field of obj…