1 code implementation • 23 Nov 2022 • Konstantin Kobs, Michael Steininger, Andreas Hotho
Therefore, we present Language-Guided Zero-Shot Deep Metric Learning (LanZ-DML) as a new DML setting in which users control the properties that should be important for image representations without training data by only using natural language.
no code implementations • 7 Jul 2022 • Padraig Davidson, Michael Steininger, André Huhn, Anna Krause, Andreas Hotho
Due to its chaotic nature and massive amounts of datapoints, timeseries are hard to label manually.
1 code implementation • ICCV 2021 • Konstantin Kobs, Michael Steininger, Andrzej Dulny, Andreas Hotho
In this paper, we investigate this by conducting a two-step analysis to extract and compare the learned visual features of the same model architecture trained with different loss functions: First, we compare the learned features on the pixel level by correlating saliency maps of the same input images.
no code implementations • 8 Oct 2021 • Padraig Davidson, Michael Steininger, Florian Lautenschlager, Anna Krause, Andreas Hotho
Sensor-equipped beehives allow monitoring the living conditions of bees.
no code implementations • 9 Dec 2020 • Michael Steininger, Daniel Abel, Katrin Ziegler, Anna Krause, Heiko Paeth, Andreas Hotho
Climate models are an important tool for the assessment of prospective climate change effects but they suffer from systematic and representation errors, especially for precipitation.
no code implementations • 10 Mar 2020 • Padraig Davidson, Michael Steininger, Florian Lautenschlager, Konstantin Kobs, Anna Krause, Andreas Hotho
Precision beekeeping allows to monitor bees' living conditions by equipping beehives with sensors.
1 code implementation • 6 Mar 2020 • Konstantin Kobs, Michael Steininger, Albin Zehe, Florian Lautenschlager, Andreas Hotho
One common loss function in neural network classification tasks is Categorical Cross Entropy (CCE), which punishes all misclassifications equally.
no code implementations • 18 Feb 2020 • Michael Steininger, Konstantin Kobs, Albin Zehe, Florian Lautenschlager, Martin Becker, Andreas Hotho
In this paper, we advocate a paradigm shift for LUR models: We propose the Data-driven, Open, Global (DOG) paradigm that entails models based on purely data-driven approaches using only openly and globally available data.