F
Frerk Saxen
Researcher at Otto-von-Guericke University Magdeburg
Publications - 22
Citations - 296
Frerk Saxen is an academic researcher from Otto-von-Guericke University Magdeburg. The author has contributed to research in topics: Facial expression & Computer science. The author has an hindex of 8, co-authored 20 publications receiving 203 citations.
Papers
More filters
Proceedings ArticleDOI
FPGA-Based Real-Time Pedestrian Detection on High-Resolution Images
TL;DR: The Histograms of Oriented Gradients descriptor is used in combination with a Support Vector Machine for classification as a basic method to process image data at twice the pixel frequency and to normalize blocks with the L1-Sqrt-norm resulting in an efficient resource utilization.
Proceedings ArticleDOI
Face Attribute Detection with MobileNetV2 and NasNet-Mobile
TL;DR: A straight forward and fast face alignment technique for preprocessing and estimate the face attributes using MobileNetV2 and Nasnet-Mobile, two lightweight CNN (Convolutional Neural Network) architectures that perform similarly well in terms of accuracy and speed.
Proceedings ArticleDOI
Handling Data Imbalance in Automatic Facial Action Intensity Estimation.
TL;DR: A novel multiclass under-sampling method is proposed for AU intensity estimation and its use in an ensemble to handle the inherent class imbalance and a comparison to state of the art methods is compared.
Proceedings ArticleDOI
Cross-Database Evaluation of Pain Recognition from Facial Video
TL;DR: This paper proposes two distinct methods to classify based on the temporal information of pain, and does cross-database validations on two benchmark pain databases: BioVid and X-ITE.
Proceedings ArticleDOI
Landmark based head pose estimation benchmark and method
TL;DR: A new database is introduced, a new benchmark protocol is proposed, and a method to learn a pose estimator on top of any landmark detector (called HPFL) is described and implemented, revealing that OpenFace comes with the best pose estimators.