| 2025 | ESANN | Investigating the Impact of Imbalanced Medical Data on the Performance of Self-Supervised Learning Approaches. | Manuel Laufer, Felicitas Brokmann, Dominik Mairhfer, Erhardt Barth, Thomas Martinetz |
| 2024 | ESANN | AI-based Collimation Optimization for X-Ray Imaging using Time-of-Flight Cameras. | Dominik Mairhfer, Manuel Laufer, Lennart Berkel, Arpad Bischof, Erhardt Barth, Jrg Barkhausen, Thomas Martinetz |
| 2024 | ICANN | Video Understanding Using 2D-CNNs on Salient Spatio-Temporal Slices. | Yaxin Hu, Erhardt Barth |
| 2024 | ICONIP | How to Efficiently Use Color and Temporal Information for Video Understanding. | Yaxin Hu, Erhardt Barth |
| 2024 | IJCNN | Novel Design Ideas that Improve Video-Understanding Networks with Transformers. | Yaxin Hu, Erhardt Barth |
| 2024 | IJCNN | Leaky ReLUs That Differ in Forward and Backward Pass Facilitate Activation Maximization in Deep Neural Networks. | Christoph Linse, Erhardt Barth, Thomas Martinetz |
| 2023 | IJCNN | Connections Between Pairs of Filters Improve the Accuracy of Convolutional Neural Networks. | Kathleen Anderson, Philipp Grning, Erhardt Barth |
| 2023 | IJCNN | Convolutional Neural Networks Do Work with Pre-Defined Filters. | Christoph Linse, Erhardt Barth, Thomas Martinetz |
| 2020 | ICANN | Log-Nets: Logarithmic Feature-Product Layers Yield More Compact Networks. | Philipp Grning, Thomas Martinetz, Erhardt Barth |
| 2017 | ACIVS | Sensing Forest for Pattern Recognition. | Irina Burciu, Thomas Martinetz, Erhardt Barth |
| 2017 | EMNLP | A Hybrid Convolutional Variational Autoencoder for Text Generation. | Stanislau Semeniuta, Aliaksei Severyn, Erhardt Barth |
| 2017 | IJCNN | Recursive autoconvolution for unsupervised learning of convolutional neural networks. | Boris Knyazev, Erhardt Barth, Thomas Martinetz |
| 2016 | COLING | Recurrent Dropout without Memory Loss. | Stanislau Semeniuta, Aliaksei Severyn, Erhardt Barth |
| 2015 | IJCNN | Deep convolutional neural networks as generic feature extractors. | Lars Hertel, Erhardt Barth, Thomas Kster, Thomas Martinetz |
| 2015 | IJCNN | Learning orthogonal sparse representations by using geodesic flow optimization. | Henry Schtze, Erhardt Barth, Thomas Martinetz |
| 2010 | ETRA | Space-variant spatio-temporal filtering of video for gaze visualization and perceptual learning. | Michael Dorr, Halszka Jarodzka, Erhardt Barth |
| 2010 | ICANN | A Learned Saliency Predictor for Dynamic Natural Scenes. | Eleonora Vig, Michael Dorr, Thomas Martinetz, Erhardt Barth |
| 2009 | ICANN | Multimodal Sparse Features for Object Detection. | Martin Haker, Thomas Martinetz, Erhardt Barth |
| 2009 | IPAS | Active contours that grow and compete driven by local region descriptors. | Cristina Darolti, Erhardt Barth, Christoph Bodensteiner, Ulrich G. Hofmann |
| 2008 | CVPR | Shading constraint improves accuracy of time-of-flight measurements. | Martin Bhme, Martin Haker, Thomas Martinetz, Erhardt Barth |
| 2008 | CVPR | Scale-invariant range features for time-of-flight camera applications. | Martin Haker, Martin Bhme, Thomas Martinetz, Erhardt Barth |
| 2008 | CVPR | ToF-sensors: New dimensions for realism and interactivity. | Andreas Kolb, Erhardt Barth, Reinhard Koch |
| 2008 | ESANN | Learning Data Representations with Sparse Coding Neural Gas. | Kai Labusch, Erhardt Barth, Thomas Martinetz |
| 2008 | ETRA | A software framework for simulating eye trackers. | Martin Bhme, Michael Dorr, Mathis Graw, Thomas Martinetz, Erhardt Barth |
| 2008 | ICANN | Sparse Coding Neural Gas for the Separation of Noisy Overcomplete Sources. | Kai Labusch, Erhardt Barth, Thomas Martinetz |
| 2006 | ETRA | Gaze-contingent temporal filtering of video. | Martin Bhme, Michael Dorr, Thomas Martinetz, Erhardt Barth |
| 2004 | GI | Saliency Extraction for Gaze-Contingent Displays. | Martin Bhme, Christopher Krause, Thomas Martinetz, Erhardt Barth |
| 2004 | ICASSP | Estimation of multiple local orientations in image signals. | Til Aach, Ingo Stuke, Cicero Mota, Erhardt Barth |
| 2004 | ICIP | Estimation of multiple orientations in multi-dimensional signals. | Cicero Mota, Til Aach, Ingo Stuke, Erhardt Barth |
| 2004 | VCIP | Estimation of multiple motions using block matching and Markov random fields. | Ingo Stuke, Til Aach, Erhardt Barth, Cicero Mota |
| 2003 | ICASSP | Linear and regularized solutions for multiple motions. | Ingo Stuke, Til Aach, Cicero Mota, Erhardt Barth |
| 2003 | ICIP | Spatio-temporal motion estimation for transparency and occlusions. | Ingo Stuke, Til Aach, Erhardt Barth, Cicero Mota |
| 2003 | SNPD | Categorization of Transparent-Motion Patterns Using the Projective Plane. | Cicero Mota, Michael Dorr, Ingo Stuke, Erhardt Barth |
| 2003 | SNPD | Estimation of Multiple Motions by Block Matching. | Ingo Stuke, Til Aach, Erhardt Barth, Cicero Mota |
| 2001 | ICIP | Analytic solutions for multiple motions. | Cicero Mota, Ingo Stuke, Erhardt Barth |