1 code implementation • EMNLP (LaTeCHCLfL, CLFL, LaTeCH) 2021 • Felix Schneider, Björn Barz, Phillip Brandes, Sophie Marshall, Joachim Denzler
In contrast, we propose an approach targeting the more general and challenging case A B B’ A’, where the words A, A’ and B, B’ constituting the chiasmus do not need to be identical but just related in meaning.
1 code implementation • LREC (MWE) 2022 • Felix Schneider, Sven Sickert, Phillip Brandes, Sophie Marshall, Joachim Denzler
We train this method in a zero-shot pseudo-supervised manner by generating artificial metaphor examples and show that our approach can be used to generate a metaphor dataset with low annotation cost.
no code implementations • ECCV 2020 • Christian Reimers, Jakob Runge, Joachim Denzler
Deep neural networks are tremendously successful in many applications, but end-to-end trained networks often result in hard to understand black-box classifiers or predictors.
no code implementations • 27 Mar 2025 • Sai Karthikeya Vemuri, Tim Büchner, Joachim Denzler
Implicit Neural Representation (INR) has emerged as a powerful tool for encoding discrete signals into continuous, differentiable functions using neural networks.
no code implementations • 21 Mar 2025 • Gideon Stein, Maha Shadaydeh, Jan Blunk, Niklas Penzel, Joachim Denzler
Causal discovery, or identifying causal relationships from observational data, is a notoriously challenging task, with numerous methods proposed to tackle it.
no code implementations • 12 Mar 2025 • Tim Büchner, Christoph Anders, Orlando Guntinas-Lichius, Joachim Denzler
We validate the effectiveness of our approach through experiments on a dataset of synchronized sEMG recordings and facial mimicry, demonstrating faithful geometry and appearance reconstruction.
no code implementations • 7 Mar 2025 • Niklas Penzel, Joachim Denzler
Our results highlight the potential of interventional explanations on the property level to reveal new insights into the behavior of deep models.
1 code implementation • 13 Jan 2025 • Ferdinand Rewicki, Joachim Denzler, Julia Niebling
Detecting and classifying abnormal system states is critical for condition monitoring, but supervised methods often fall short due to the rarity of anomalies and the lack of labeled data.
no code implementations • 27 Nov 2024 • Dong Han, Yong Li, Joachim Denzler
With the advancement of face reconstruction (FR) systems, privacy-preserving face recognition (PPFR) has gained popularity for its secure face recognition, enhanced facial privacy protection, and robustness to various attacks.
no code implementations • 23 Oct 2024 • Laines Schmalwasser, Jakob Gawlikowski, Joachim Denzler, Julia Niebling
This is done by mapping the most relevant images representing a CAV into a text-image embedding where a joint description of these relevant images can be computed.
no code implementations • 24 Sep 2024 • Tim Büchner, Niklas Penzel, Orlando Guntinas-Lichius, Joachim Denzler
Understanding expressions is vital for deciphering human behavior, and nowadays, end-to-end trained black box models achieve high performance.
1 code implementation • 23 Aug 2024 • Sai Karthikeya Vemuri, Tim Büchner, Julia Niebling, Joachim Denzler
We leverage tensor decomposition forms to separate the variables in a PINN setting.
no code implementations • 6 Jul 2024 • Dong Han, Yufan Jiang, Yong Li, Ricardo Mendes, Joachim Denzler
In this work, we leverage the pure skin color patch from the face image as the additional information to train an auxiliary skin color feature extractor and face recognition model in parallel to improve performance of state-of-the-art (SOTA) privacy-preserving face recognition (PPFR) systems.
1 code implementation • 14 Jun 2024 • Ferdinand Rewicki, Jakob Gawlikowski, Julia Niebling, Joachim Denzler
The detection of abnormal or critical system states is essential in condition monitoring.
no code implementations • 18 Apr 2024 • Tristan Piater, Niklas Penzel, Gideon Stein, Joachim Denzler
To evaluate this trend for medical imaging, we extend two widely adopted convolutional architectures with different self-attention variants on two different medical datasets.
no code implementations • 18 Apr 2024 • Niklas Penzel, Gideon Stein, Joachim Denzler
If later such a bias is discovered during inference or deployment, it is often necessary to acquire new data and retrain the model.
no code implementations • 11 Apr 2024 • Tim Büchner, Niklas Penzel, Orlando Guntinas-Lichius, Joachim Denzler
We introduce a workflow to evaluate explicit properties and their impact.
1 code implementation • 14 Feb 2024 • Gideon Stein, Maha Shadaydeh, Joachim Denzler
Our empirical findings suggest that causal discovery in a supervised manner is possible, assuming that the training and test time series samples share most of their dynamics.
1 code implementation • 13 Feb 2024 • Tim Büchner, Oliver Mothes, Orlando Guntinas-Lichius, Joachim Denzler
Given that solely the facial muscles, innervated by the facial nerve, are responsible for facial expressions, facial palsy can lead to severe impairments in facial movements.
no code implementations • 24 Jan 2024 • Dong Han, Yong Li, Joachim Denzler
Lastly, secure multiparty computation is implemented for safely computing the embedding distance during model inference.
1 code implementation • 16 Jan 2024 • Wasim Ahmad, Maha Shadaydeh, Joachim Denzler
Causal inference in a nonlinear system of multivariate timeseries is instrumental in disentangling the intricate web of relationships among variables, enabling us to make more accurate predictions and gain deeper insights into real-world complex systems.
no code implementations • 9 Nov 2023 • Tim Büchner, Sven Sickert, Gerd Fabian Volk, Christoph Anders, Orlando Guntinas-Lichius, Joachim Denzler
As a result, existing facial analysis methods can be used without further changes with respect to the data.
no code implementations • 28 Jul 2023 • Dimitri Korsch, Paul Bodesheim, Joachim Denzler
Biodiversity monitoring is crucial for tracking and counteracting adverse trends in population fluctuations.
no code implementations • 28 Jul 2023 • Dimitri Korsch, Paul Bodesheim, Gunnar Brehm, Joachim Denzler
We used this dataset to develop and evaluate a two-stage pipeline for insect detection and moth species classification in previous work.
no code implementations • 27 Jul 2023 • Dimitri Korsch, Maha Shadaydeh, Joachim Denzler
It utilizes the concrete dropout (CD) to sample a set of attribution masks and updates the sampling parameters based on the output of the classification model.
no code implementations • 17 Jul 2023 • Matthias Körschens, Solveig Franziska Bucher, Christine Römermann, Joachim Denzler
The plant community composition is an essential indicator of environmental changes and is, for this reason, usually analyzed in ecological field studies in terms of the so-called plant cover.
1 code implementation • 21 Dec 2022 • Ferdinand Rewicki, Joachim Denzler, Julia Niebling
Detecting anomalies in time series data is important in a variety of fields, including system monitoring, healthcare, and cybersecurity.
Ranked #5 on
Anomaly Detection
on UCR Anomaly Archive
no code implementations • 23 Sep 2022 • Violeta Teodora Trifunov, Maha Shadaydeh, Joachim Denzler
We compare our results on synthetic data to those of a time series deconfounding method both with and without estimated confounders.
1 code implementation • 8 Jul 2022 • Wasim Ahmad, Maha Shadaydeh, Joachim Denzler
Cause-effect analysis is crucial to understand the underlying mechanism of a system.
1 code implementation • IEEE Access 2022 • Lorenzo Brigato, Björn Barz, Luca Iocchi, Joachim Denzler
However, as research in this scope is still in its infancy, two key ingredients are missing for ensuring reliable and truthful progress: a systematic and extensive overview of the state of the art, and a common benchmark to allow for objective comparisons between published methods.
no code implementations • 22 Oct 2021 • Bernd Gruner, Matthias Körschens, Björn Barz, Joachim Denzler
We discovered that domain adaptation works very well for fine-grained recognition and that the normalization methods have a great influence on the results.
no code implementations • 28 Sep 2021 • Björn Barz, Lorenzo Brigato, Luca Iocchi, Joachim Denzler
Learning from limited amounts of data is the hallmark of intelligence, requiring strong generalization and abstraction skills.
no code implementations • 22 Sep 2021 • Wasim Ahmad, Maha Shadaydeh, Joachim Denzler
Overall our method outperforms the widely used vector autoregressive Granger causality and PCMCI in detecting nonlinear causal dependency in multivariate time series.
no code implementations • 14 Sep 2021 • Violeta Teodora Trifunov, Maha Shadaydeh, Björn Barz, Joachim Denzler
There are numerous methods for detecting anomalies in time series, but that is only the first step to understanding them.
1 code implementation • 30 Aug 2021 • Lorenzo Brigato, Björn Barz, Luca Iocchi, Joachim Denzler
Data-efficient image classification using deep neural networks in settings, where only small amounts of labeled data are available, has been an active research area in the recent past.
Ranked #1 on
Small Data Image Classification
on DEIC Benchmark
no code implementations • 16 Aug 2021 • Björn Barz, Joachim Denzler
We introduce a novel dataset for architectural style classification, consisting of 9, 485 images of church buildings.
no code implementations • 21 Jun 2021 • Matthias Körschens, Paul Bodesheim, Christine Römermann, Solveig Franziska Bucher, Mirco Migliavacca, Josephine Ulrich, Joachim Denzler
Monitoring the responses of plants to environmental changes is essential for plant biodiversity research.
no code implementations • 22 Apr 2021 • Clemens-Alexander Brust, Björn Barz, Joachim Denzler
Learning from imprecise labels such as "animal" or "bird", but making precise predictions like "snow bunting" at inference time is an important capability for any classifier when expertly labeled training data is scarce.
2 code implementations • 16 Apr 2021 • Christian Requena-Mesa, Vitus Benson, Markus Reichstein, Jakob Runge, Joachim Denzler
We frame Earth surface forecasting as the task of predicting satellite imagery conditioned on future weather.
no code implementations • 10 Mar 2021 • Christian Reimers, Paul Bodesheim, Jakob Runge, Joachim Denzler
Often, the bias of a classifier is a direct consequence of a bias in the training dataset, frequently caused by the co-occurrence of relevant features and irrelevant ones.
no code implementations • 1 Feb 2021 • Oana-Iuliana Popescu, Maha Shadaydeh, Joachim Denzler
Heuristic methods result in false-positive artifacts because the image after the perturbation is far from the original data distribution.
no code implementations • 16 Dec 2020 • Maha Shadaydeh, Lea Mueller, Dana Schneider, Martin Thuemmel, Thomas Kessler, Joachim Denzler
Identifying the direction of emotional influence in a dyadic dialogue is of increasing interest in the psychological sciences with applications in psychotherapy, analysis of political interactions, or interpersonal conflict behavior.
1 code implementation • 11 Dec 2020 • Christian Requena-Mesa, Vitus Benson, Joachim Denzler, Jakob Runge, Markus Reichstein
Here, we define high-resolution Earth surface forecasting as video prediction of satellite imagery conditional on mesoscale weather forecasts.
no code implementations • 12 Nov 2020 • Björn Barz, Joachim Denzler
Content-based image retrieval has seen astonishing progress over the past decade, especially for the task of retrieving images of the same object that is depicted in the query image.
1 code implementation • 11 Nov 2020 • Björn Barz, Kai Schröter, Ann-Christin Kra, Joachim Denzler
The analysis of natural disasters such as floods in a timely manner often suffers from limited data due to coarsely distributed sensors or sensor failures.
no code implementations • 13 Oct 2020 • Clemens-Alexander Brust, Björn Barz, Joachim Denzler
For example, a non-breeding snow bunting is labeled as a bird.
1 code implementation • 4 Jul 2020 • Dimitri Korsch, Paul Bodesheim, Joachim Denzler
We assume that part-based methods suffer from a missing representation of local features, which is invariant to the order of parts and can handle a varying number of visible parts appropriately.
Ranked #16 on
Fine-Grained Image Classification
on NABirds
no code implementations • 23 Jun 2020 • Nils Gählert, Jun-Jun Wan, Nicolas Jourdan, Jan Finkbeiner, Uwe Franke, Joachim Denzler
In this paper we propose a novel 3D single-shot object detection method for detecting vehicles in monocular RGB images.
no code implementations • 15 Jun 2020 • Nils Gählert, Niklas Hanselmann, Uwe Franke, Joachim Denzler
Object detection is an important task in environment perception for autonomous driving.
1 code implementation • 14 Jun 2020 • Nils Gählert, Nicolas Jourdan, Marius Cordts, Uwe Franke, Joachim Denzler
In addition, we complement the Cityscapes benchmark suite with 3D vehicle detection based on the new annotations as well as metrics presented in this work.
no code implementations • 20 Jan 2020 • Clemens-Alexander Brust, Christoph Käding, Joachim Denzler
By selecting unlabeled examples that are promising in terms of model improvement and only asking for respective labels, active learning can increase the efficiency of the labeling process in terms of time and cost.
2 code implementations • 16 Oct 2019 • Jonas Jäger, Gereon Reus, Joachim Denzler, Viviane Wolff, Klaus Fricke-Neuderth
In this paper we present a framework that allows for a quick and flexible design of semi-automatic annotation pipelines.
1 code implementation • 14 Oct 2019 • Martin Thümmel, Sven Sickert, Joachim Denzler
As a countermeasure, we propose a method that automatically analyzes the plausibility of facial behavior based on a sequence of 3D face scans.
no code implementations • 23 Sep 2019 • Christian Requena-Mesa, Markus Reichstein, Miguel Mahecha, Basil Kraft, Joachim Denzler
We demonstrate that for many purposes the generated landscapes behave as real with immediate application for global change studies.
2 code implementations • 16 Sep 2019 • Dimitri Korsch, Paul Bodesheim, Joachim Denzler
Fine-grained visual categorization is a classification task for distinguishing categories with high intra-class and small inter-class variance.
Ranked #20 on
Fine-Grained Image Classification
on NABirds
no code implementations • 12 Aug 2019 • Oliver Mothes, Joachim Denzler
In combination with synthetically generated character data, the real data is used to train efficient convolutional neural networks for character classification serving a practical runtime as well as a high accuracy.
Optical Character Recognition
Optical Character Recognition (OCR)
1 code implementation • 9 Aug 2019 • Björn Barz, Kai Schröter, Moritz Münch, Bin Yang, Andrea Unger, Doris Dransch, Joachim Denzler
The analysis of natural disasters such as floods in a timely manner often suffers from limited data due to a coarse distribution of sensors or sensor failures.
no code implementations • 1 Feb 2019 • Björn Barz, Joachim Denzler
However, we find that 3. 3% and 10% of the images from the test sets of these datasets have duplicates in the training set.
1 code implementation • 25 Jan 2019 • Björn Barz, Joachim Denzler
The categorical cross-entropy loss after softmax activation is the method of choice for classification.
no code implementations • 11 Dec 2018 • Matthias Körschens, Björn Barz, Joachim Denzler
Identifying animals from a large group of possible individuals is very important for biodiversity monitoring and especially for collecting data on a small number of particularly interesting individuals, as these have to be identified first before this can be done.
no code implementations • 17 Nov 2018 • Clemens-Alexander Brust, Joachim Denzler
In this paper, we propose to make use of preexisting class hierarchies like WordNet to integrate additional domain knowledge into classification.
no code implementations • 17 Nov 2018 • Clemens-Alexander Brust, Joachim Denzler
In this paper, we use five different semantic and visual similarity measures each to thoroughly analyze the relationship without relying too much on any single definition.
no code implementations • 29 Oct 2018 • Lea Müller, Maha Shadaydeh, Martin Thümmel, Thomas Kessler, Dana Schneider, Joachim Denzler
Human nonverbal emotional communication in dyadic dialogs is a process of mutual influence and adaptation.
no code implementations • 26 Sep 2018 • Clemens-Alexander Brust, Christoph Käding, Joachim Denzler
In this paper, we combine a novel method of active learning for object detection with an incremental learning scheme to enable continuous exploration of new unlabeled datasets.
1 code implementation • 26 Sep 2018 • Björn Barz, Joachim Denzler
Such an embedding does not only improve image retrieval results, but could also facilitate integrating semantics for other tasks, e. g., novelty detection or few-shot learning.
1 code implementation • 7 Sep 2018 • Björn Barz, Christoph Käding, Joachim Denzler
We propose Information-Theoretic Active Learning (ITAL), a novel batch-mode active learning method for binary classification, and apply it for acquiring meaningful user feedback in the context of content-based image retrieval.
1 code implementation • 19 Apr 2018 • Björn Barz, Erik Rodner, Yanira Guanche Garcia, Joachim Denzler
Automatic detection of anomalies in space- and time-varying measurements is an important tool in several fields, e. g., fraud detection, climate analysis, or healthcare monitoring.
1 code implementation • 2 Nov 2017 • Björn Barz, Joachim Denzler
Query images presented to content-based image retrieval systems often have various different interpretations, making it difficult to identify the search objective pursued by the user.
no code implementations • 23 May 2017 • Talha Qaiser, Abhik Mukherjee, Chaitanya Reddy Pb, Sai Dileep Munugoti, Vamsi Tallam, Tomi Pitkäaho, Taina Lehtimäki, Thomas Naughton, Matt Berseth, Aníbal Pedraza, Ramakrishnan Mukundan, Matthew Smith, Abhir Bhalerao, Erik Rodner, Marcel Simon, Joachim Denzler, Chao-Hui Huang, Gloria Bueno, David Snead, Ian Ellis, Mohammad Ilyas, Nasir Rajpoot
In this paper, we report on a recent automated Her2 scoring contest, held in conjunction with the annual PathSoc meeting held in Nottingham in June 2016, aimed at systematically comparing and advancing the state-of-the-art Artificial Intelligence (AI) based automated methods for Her2 scoring.
2 code implementations • ICCV 2017 • Marcel Simon, Yang Gao, Trevor Darrell, Joachim Denzler, Erik Rodner
In this paper, we generalize average and bilinear pooling to "alpha-pooling", allowing for learning the pooling strategy during training.
no code implementations • 10 Apr 2017 • Björn Barz, Erik Rodner, Christoph Käding, Joachim Denzler
We combine features extracted from pre-trained convolutional neural networks (CNNs) with the fast, linear Exemplar-LDA classifier to get the advantages of both: the high detection performance of CNNs, automatic feature engineering, fast model learning from few training samples and efficient sliding-window detection.
no code implementations • 5 Mar 2017 • Marc Aubreville, Christian Knipfer, Nicolai Oetter, Christian Jaremenko, Erik Rodner, Joachim Denzler, Christopher Bohr, Helmut Neumann, Florian Stelzle, Andreas Maier
For this work, CLE image sequences (7894 images) from patients diagnosed with OSCC were obtained from 4 specific locations in the oral cavity, including the OSCC lesion.
no code implementations • 19 Dec 2016 • Christoph Käding, Erik Rodner, Alexander Freytag, Joachim Denzler
The demands on visual recognition systems do not end with the complexity offered by current large-scale image datasets, such as ImageNet.
4 code implementations • 5 Dec 2016 • Marcel Simon, Erik Rodner, Joachim Denzler
Convolutional neural networks (CNN) pre-trained on ImageNet are the backbone of most state-of-the-art approaches.
no code implementations • 21 Oct 2016 • Erik Rodner, Björn Barz, Yanira Guanche, Milan Flach, Miguel Mahecha, Paul Bodesheim, Markus Reichstein, Joachim Denzler
We present new methods for batch anomaly detection in multivariate time series.
no code implementations • 21 Oct 2016 • Erik Rodner, Marcel Simon, Robert B. Fisher, Joachim Denzler
In this paper, we study the sensitivity of CNN outputs with respect to image transformations and noise in the area of fine-grained recognition.
no code implementations • 10 Oct 2016 • Manuel Amthor, Erik Rodner, Joachim Denzler
We propose Impatient Deep Neural Networks (DNNs) which deal with dynamic time budgets during application.
no code implementations • 19 Sep 2016 • Seyed Ali Amirshahi, Gregor Uwe Hayn-Leichsenring, Joachim Denzler, Christoph Redies
As a first step and to assess this dataset, using a classifier, we investigate the correlation between the subjective scores and two widely used features that are related to color perception and in different aesthetic quality assessment approaches.
no code implementations • 14 Jun 2016 • Clemens-Alexander Brust, Sven Sickert, Marcel Simon, Erik Rodner, Joachim Denzler
Neural networks and especially convolutional neural networks are of great interest in current computer vision research.
no code implementations • 3 Jul 2015 • Erik Rodner, Marcel Simon, Gunnar Brehm, Stephanie Pietsch, J. Wolfgang Wägele, Joachim Denzler
In the following paper, we present and discuss challenging applications for fine-grained visual classification (FGVC): biodiversity and species analysis.
no code implementations • CVPR 2015 • Christoph Kading, Alexander Freytag, Erik Rodner, Paul Bodesheim, Joachim Denzler
In active learning, all categories occurring in collected data are usually assumed to be known in advance and experts should be able to label every requested instance.
no code implementations • 23 Feb 2015 • Clemens-Alexander Brust, Sven Sickert, Marcel Simon, Erik Rodner, Joachim Denzler
Classifying single image patches is important in many different applications, such as road detection or scene understanding.
1 code implementation • 12 Nov 2014 • Marcel Simon, Erik Rodner, Joachim Denzler
Current fine-grained classification approaches often rely on a robust localization of object parts to extract localized feature representations suitable for discrimination.
no code implementations • 20 Aug 2014 • Alexander Freytag, Johannes Rühle, Paul Bodesheim, Erik Rodner, Joachim Denzler
To answer this question, we present an in-depth analysis of the effect of local feature quantization on human recognition performance.
no code implementations • 10 Jul 2014 • Björn Barz, Erik Rodner, Joachim Denzler
ARTOS is all about creating, tuning, and applying object detection models with just a few clicks.
no code implementations • CVPR 2014 • Daniel Haase, Erik Rodner, Joachim Denzler
Therefore, we present a transfer learning method that is able to learn from related training data using an instance-weighted transfer technique.
no code implementations • CVPR 2014 • Christoph Goring, Erik Rodner, Alexander Freytag, Joachim Denzler
In the following paper, we present an approach for finegrained recognition based on a new part detection method.
no code implementations • 17 Oct 2013 • Christoph Göring, Alexander Freytag, Erik Rodner, Joachim Denzler
In this paper, we tackle the problem of visual categorization of dog breeds, which is a surprisingly challenging task due to simultaneously present low interclass distances and high intra-class variances.
no code implementations • CVPR 2013 • Paul Bodesheim, Alexander Freytag, Erik Rodner, Michael Kemmler, Joachim Denzler
In contrast to modeling the support of each known class individually, our approach makes use of a projection in a joint subspace where training samples of all known classes have zero intra-class variance.