Get our free extension to see links to code for papers anywhere online!

Add to Chrome

Add to Firefox

Picture for Tobias Riedlinger

Tobias Riedlinger

Pixel-wise Gradient Uncertainty for Convolutional Neural Networks applied to Out-of-Distribution Segmentation

Mar 13, 2023
Kira Maag, Tobias Riedlinger

Figure 1 for Pixel-wise Gradient Uncertainty for Convolutional Neural Networks applied to Out-of-Distribution Segmentation

Figure 2 for Pixel-wise Gradient Uncertainty for Convolutional Neural Networks applied to Out-of-Distribution Segmentation

Figure 3 for Pixel-wise Gradient Uncertainty for Convolutional Neural Networks applied to Out-of-Distribution Segmentation

Figure 4 for Pixel-wise Gradient Uncertainty for Convolutional Neural Networks applied to Out-of-Distribution Segmentation

Access Paper or Ask Questions

Towards Rapid Prototyping and Comparability in Active Learning for Deep Object Detection

Dec 21, 2022
Tobias Riedlinger, Marius Schubert, Karsten Kahl, Hanno Gottschalk, Matthias Rottmann

Figure 1 for Towards Rapid Prototyping and Comparability in Active Learning for Deep Object Detection

Figure 2 for Towards Rapid Prototyping and Comparability in Active Learning for Deep Object Detection

Figure 3 for Towards Rapid Prototyping and Comparability in Active Learning for Deep Object Detection

Figure 4 for Towards Rapid Prototyping and Comparability in Active Learning for Deep Object Detection

* 17 pages, 12 figures, 9 tables

Access Paper or Ask Questions

Gradient-Based Quantification of Epistemic Uncertainty for Deep Object Detectors

Jul 09, 2021
Tobias Riedlinger, Matthias Rottmann, Marius Schubert, Hanno Gottschalk

Figure 1 for Gradient-Based Quantification of Epistemic Uncertainty for Deep Object Detectors

Figure 2 for Gradient-Based Quantification of Epistemic Uncertainty for Deep Object Detectors

Figure 3 for Gradient-Based Quantification of Epistemic Uncertainty for Deep Object Detectors

Figure 4 for Gradient-Based Quantification of Epistemic Uncertainty for Deep Object Detectors

* 20 pages, 8 figures, 7 tables

Access Paper or Ask Questions