Aysim Toker
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Aysim Toker has a total of 8 co-authors. Here are the top co-authors by number of shared publications:
- Laura Leal-Taixé
Dynamic Vision and Learning Group, Technical University of Munich, Munich, Germany
2 shared publications - Bernhard Schölkopf
ETH Zurich
1 shared publication Daniel Cremers
1 shared publication- Florian Bernard
Service de Neurochirurgie, CHU d'Angers, Angers, France; Laboratoire d'Anatomie, Faculté de Médecine d'Angers, Angers, France
1 shared publication - Lukas Kondmann
Remote Sensing Technology Institute, German Aerospace Center (DLR), Wessling, Germany
1 shared publication Marvin Eisenberger
1 shared publication
Latest Publications
- Open Access
A Unified Framework for Implicit Sinkhorn Differentiation
The Sinkhorn operator has recently experienced a surge of popularity in computer vision and related fields. One major reason is its ease of integration into deep learning frameworks. To allow for an efficient training of respective neural networks, we propose an algorithm that obtains analytical gradients of a Sinkhorn layer via implicit differentiation. In comparison to prior work, our framework is based on the most general formulation of the Sinkhorn operator. It allows for any type of loss function, while both the target capacities and cost matrices are differentiated jointly. We further construct error bounds of the resulting algorithm for approximate inputs. Finally, we demonstrate that for a number of applications, simply replacing automatic differentiation with our algorithm directly improves the stability and accuracy of the obtained gradients. Moreover, we show that it is computationally more efficient, particularly when resources like GPU memory are scarce.
Open AccessA Unified Framework for Implicit Sinkhorn Differentiation
The Sinkhorn operator has recently experienced a surge of popularity in computer vision and related fields. One major reason is its ease of integration into deep learning frameworks. To allow for an efficient training of respective neural networks, we propose an algorithm that obtains analytical gradients of a Sinkhorn layer via implicit differentiation. In comparison to prior work, our framework is based on the most general formulation of the Sinkhorn operator. It allows for any type of loss function, while both the target capacities and cost matrices are differentiated jointly. We further construct error bounds of the resulting algorithm for approximate inputs. Finally, we demonstrate that for a number of applications, simply replacing automatic differentiation with our algorithm directly improves the stability and accuracy of the obtained gradients. Moreover, we show that it is computationally more efficient, particularly when resources like GPU memory are scarce.
- Open Access
- 25 November 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 60 (01962892) , 1-15
Spatial Context Awareness for Unsupervised Change Detection in Optical Satellite Images
Detecting changes on the ground in multitemporal Earth observation data is one of the key problems in remote sensing. In this article, we introduce Sibling Regression for Optical Change detection (SiROC), an unsupervised method for change detection (CD) in optical satellite images with medium and high resolutions. SiROC is a spatial context-based method that models a pixel as a linear combination of its distant neighbors. It uses this model to analyze differences in the pixel and its spatial context-based predictions in subsequent time periods for CD. We combine this spatial context-based CD with ensembling over mutually exclusive neighborhoods and transitioning from pixel to object-level changes with morphological operations. SiROC achieves competitive performance for CD with medium-resolution Sentinel-2 and high-resolution Planetscope imagery on four datasets. Besides accurate predictions without the need for training, SiROC also provides a well-calibrated uncertainty of its predictions.
- 25 November 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 60 (01962892) , 1-15
Open AccessSpatial Context Awareness for Unsupervised Change Detection in Optical Satellite Images
Detecting changes on the ground in multitemporal Earth observation data is one of the key problems in remote sensing. In this article, we introduce Sibling Regression for Optical Change detection (SiROC), an unsupervised method for change detection (CD) in optical satellite images with medium and high resolutions. SiROC is a spatial context-based method that models a pixel as a linear combination of its distant neighbors. It uses this model to analyze differences in the pixel and its spatial context-based predictions in subsequent time periods for CD. We combine this spatial context-based CD with ensembling over mutually exclusive neighborhoods and transitioning from pixel to object-level changes with morphological operations. SiROC achieves competitive performance for CD with medium-resolution Sentinel-2 and high-resolution Planetscope imagery on four datasets. Besides accurate predictions without the need for training, SiROC also provides a well-calibrated uncertainty of its predictions.