O
Olaf Hellwich
Researcher at Technical University of Berlin
Publications - 213
Citations - 2672
Olaf Hellwich is an academic researcher from Technical University of Berlin. The author has contributed to research in topics: Synthetic aperture radar & Image segmentation. The author has an hindex of 26, co-authored 191 publications receiving 2071 citations. Previous affiliations of Olaf Hellwich include Free University of Berlin & Technische Universität München.
Papers
More filters
Journal ArticleDOI
Deep convolutional neural networks for automatic classification of gastric carcinoma using whole slide images in digital histopathology
TL;DR: This study explores deep learning methods for computer-aided classification in H&E stained histopathological whole slide images of gastric carcinoma using the widely known AlexNet deep convolutional framework.
Journal ArticleDOI
Towards Global Volcano Monitoring Using Multisensor Sentinel Missions and Artificial Intelligence: The MOUNTS Monitoring System
Sébastien Valade,Andreas Ley,Francesco Massimetti,Olivier D'Hondt,Marco Laiolo,Diego Coppola,David Loibl,Olaf Hellwich,Thomas R. Walter +8 more
TL;DR: The volcano monitoring platform MOUNTS (Monitoring Unrest from Space), which aims for global monitoring, is presented, using multisensor satellite-based imagery, ground-based seismic data, and artificial intelligence (AI) to assist monitoring tasks and it is demonstrated that AI can play a key role in such monitoring frameworks.
Journal ArticleDOI
Comparison of 3d interest point detectors and descriptors for point cloud fusion
TL;DR: The benefits and limitations of keypoints for the task of fusing multiple 3D point clouds are discussed and it is indicated that the specific method to extract and describe keypoints in 3D data has to be carefully chosen.
Journal ArticleDOI
Deep Learning Models for Retinal Blood Vessels Segmentation: A Review
Toufique Ahmed Soomro,Ahmed J. Afifi,Lihong Zheng,Shafiullah Soomro,Junbin Gao,Olaf Hellwich,Manoranjan Paul +6 more
TL;DR: This paper characterizes each deep learning based segmentation method as described in the literature, along with the limitations and advantages of each method, and offers some recommendations for future improvement for retinal image analysis.
Proceedings Article
Head pose estimation in face recognition across pose scenarios
M. Saquib Sarfraz,Olaf Hellwich +1 more
TL;DR: A novel discriminative feature description that encodes underlying shape well and is insensitive to illumination and other common variations in facial appearance, such as skin colour etc., is proposed and a pose similarity feature space (PSFS) is generated that turns the multi-class problem into two-class by using inter-pose and intra-pose similarities.