Personal information

Verified email domains

Activities

Works (13)

Pixel-Wise Gradient Uncertainty for Convolutional Neural Networks Applied to Out-of-Distribution Segmentation

Proceedings of the 19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications
2024 | Conference paper
Contributors: Kira Maag; Tobias Riedlinger
Source: Self-asserted source
Kira Maag

Uncertainty-Based Detection of Adversarial Attacks in Semantic Segmentation

Proceedings of the 19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications
2024 | Conference paper
Contributors: Kira Maag; Asja Fischer
Source: Self-asserted source
Kira Maag

Reducing Texture Bias of Deep Neural Networks via Edge Enhancing Diffusion

Frontiers in Artificial Intelligence and Applications
2024-10-16 | Conference paper
Part of ISBN: 9781643685489
Part of ISSN: 0922-6389
Part of ISSN: 1879-8314
Contributors: Edgar Heinert; Matthias Rottmann; Kira Maag; Karsten Kahl
Source: Self-asserted source
Kira Maag

Uncertainty-weighted Loss Functions for Improved Adversarial Attacks on Semantic Segmentation

2024 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)
2024-01-03 | Conference paper
Contributors: Kira Maag; Asja Fischer
Source: Self-asserted source
Kira Maag

False Negative Reduction in Semantic Segmentation Under Domain Shift Using Depth Estimation

Proceedings of the 18th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications
2023 | Conference paper
Contributors: Kira Maag; Matthias Rottmann
Source: Self-asserted source
Kira Maag

Two Video Data Sets for Tracking and Retrieval of Out of Distribution Objects

Computer Vision – ACCV 2022
2023 | Conference paper
Part of ISBN: 9783031263477
Part of ISBN: 9783031263484
Part of ISSN: 0302-9743
Part of ISSN: 1611-3349
Contributors: Kira Maag; Robin Chan; Svenja Uhlemeyer; Kamil Kowol; Hanno Gottschalk
Source: Self-asserted source
Kira Maag

Detection of Iterative Adversarial Attacks via Counter Attack

Journal of Optimization Theory and Applications
2023-08-04 | Journal article
Part of ISSN: 0022-3239
Part of ISSN: 1573-2878
Contributors: Matthias Rottmann; Kira Maag; Mathis Peyron; Hanno Gottschalk; Nataša Krejić
Source: Self-asserted source
Kira Maag
grade
Preferred source (of 2)‎

Prediction Rating and Performance Improvement for Segmentation Networks by Time-Dynamic Uncertainty Estimates

Bergische Universität Wuppertal
2021 | Dissertation or Thesis
Contributors: Kira Maag
Source: Self-asserted source
Kira Maag

False Negative Reduction in Video Instance Segmentation using Uncertainty Estimates

2021 IEEE 33rd International Conference on Tools with Artificial Intelligence (ICTAI)
2021-11 | Conference paper
Contributors: Kira Maag
Source: Self-asserted source
Kira Maag

Improving Video Instance Segmentation by Light-weight Temporal Uncertainty Estimates

2021 International Joint Conference on Neural Networks (IJCNN)
2021-07-18 | Conference paper
Contributors: Kira Maag; Matthias Rottmann; Serin Varghese; Fabian Huger; Peter Schlicht; Hanno Gottschalk
Source: Self-asserted source
Kira Maag

An Unsupervised Temporal Consistency (TC) Loss to Improve the Performance of Semantic Segmentation Networks

2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)
2021-06 | Conference paper
Contributors: Serin Varghese; Sharat Gujamagadi; Marvin Klingner; Nikhil Kapoor; Andreas Bar; Jan David Schneider; Kira Maag; Peter Schlicht; Fabian Huger; Tim Fingscheidt
Source: Self-asserted source
Kira Maag

Time-Dynamic Estimates of the Reliability of Deep Semantic Segmentation Networks

2020 IEEE 32nd International Conference on Tools with Artificial Intelligence (ICTAI)
2020-11 | Conference paper
Contributors: Kira Maag; Matthias Rottmann; Hanno Gottschalk
Source: Self-asserted source
Kira Maag

Detection of False Positive and False Negative Samples in Semantic Segmentation

2020 Design, Automation & Test in Europe Conference & Exhibition (DATE)
2020-03 | Conference paper
Contributors: Matthias Rottmann; Kira Maag; Robin Chan; Fabian Huger; Peter Schlicht; Hanno Gottschalk
Source: Self-asserted source
Kira Maag