Laslo Dinges
Wiss. Mitarbeiter/-in
Dr.-Ing. Laslo Dinges
Institut für Informations- und Kommunikationstechnik (IIKT)
Fachgebiet Neuro-Informationstechnik
Postdoc
Fachgebiet Neuro-Informationstechnik
Postdoc
Gebäude 09, Universitätsplatz 2,
39106, Magdeburg,
G09-425
Research Focus
- Pattern Recognition
- Segmentation based and holistic handwriting recognition
- Statistical model based synthetesis of handwriting samples
- Document analysis
- Optical 3D Surface Inspection
- Realtime analysis of facial expressions (in the context of human-robot interaction)
Projects
Supervision of Bachelor/Master Theses
- "Fast and Precise Binary Instance Segmentation of 2D Objects for Automotive Applications", Master Thesis at OvGU Magdeburg (practical part at CMORE AUTOMOTIVE GMBH), 2019
Lectures
- Automatic Recognition of Arabic handwriting, part of "Angewandte Bildverarbeitung" (applied image processing), apl. Prof. Dr.-Ing. habil. Al-Hamadi
- Grundlagen der Signalorientierten Bildverarbeitung (in Vertretung)
Journals
- Handrich, S.; Dinges, L.; Al-Hamadi, A; Werner, P; Saxen, F. and Al Aghbari, Z. Simultaneous prediction of valence / arousal and emotion categories and its application in an HRC scenario, Journal of Ambient Intelligence and Humanized Computing , vol. 12, no. 1, pp. 57–73, 2021.
- Dinges, L, Al-Hamadi, A.; Elzobi, M. & El-etriby, S. Synthesis of Common Arabic Handwritings to Aid Optical Character Recognition Research, Sensors , vol. 16, no. 3, p. 346, 2016.
- Dinges, L.; Al-Hamadi, A.; Elzobi, M.; El etriby, S. & Ghoneim, A. ASM based Synthesis of Handwritten Arabic Text Pages, Scientific World Journal, 2015
- Elzobi, M.; Al-Hamadi, A.; Al Aghbari, Z. & Laslo, D. IESK-ArDB: a database for handwritten Arabic and optimized topological segmentation approach International Journal on Document Analysis and Recognition, Springer, 2012
- Dinges, L.; Al-Hamadi, A.; Elzobi, M.; Al-Aghbari, Z. & Mustafa, H. Offline Automatic Segmentation based Recognition of Handwritten Arabic Words International Journal of Signal Processing, Image Processing and Pattern Recognition (IJSIP), 2011, 4, 131-144
Conference papers
-
Abdelrahman, A. A.; Hempel, T.; Khalifa, A.; Al-Hamadi, A. & Dinges, L. "L2cs-net: Fine-grained gaze estimation in unconstrained environments"
2023 8th International Conference on Frontiers of Signal Processing (ICFSP), 2023, 98-102 - Dinges, L.; Fiedler, M.-A.; Al-Hamadi, A.; Abdelrahman, A.; Weimann, J. & Bershadskyy, "Uncovering Lies: Deception Detection in a Rolling-Dice Experiment", Image Analysis and Processing -- ICIAP 2023, Springer Nature Switzerland, 2023, 293-303
- Hempel, T.; Dinges, L. & Al-Hamadi, A. Sentiment-based Engagement Strategies for intuitive Human-Robot Interaction arXiv preprint arXiv:2301.03867, VISIGRAPP (4: VISAPP), 2023
-
Ravindra, G.; D.; Dinges; L.; Al-Hamadi; A.; Baranau & V. "Fast and Precise Binary Instance Segmentation of 2D Objects for Automotive Applications" Computer Science Research Notes, Union Agency, Science Press, 3201, 302-305, 2022
- Hempel, T.; Fiedler, M.-A.; Khalifa, A.; Al-Hamadi, A. & Dinges, L. "Semantic-Aware Environment Perception for Mobile Human-Robot Interaction" 2021 12th International Symposium on Image and Signal Processing and Analysis (ISPA), 2021
- Dinges, L.; Al-Hamadi, A.; Hempel, T. & Al Aghbari, Z. Using Facial Action Recognition to Evaluate User Perception in Aggravated HRC Scenarios 2021 12th International Symposium on Image and Signal Processing and Analysis (ISPA), 2021, 195-199
- Handrich, S.; Dinges, L.; Al-Hamadi, A.; Werner, P. & Z. A. Aghbari, “Simultaneous Prediction of Valence/Arousal and Emotions on AffectNet, Aff-Wild and AFEW-VA,” in The 11th International Conference on Ambient Systems, Networks and Technologies (ANT 2020) / The 3rd International Conference on Emerging Data and Industry 4.0 (EDI40 2020) / Affiliated Workshops, April 6-9, 2020, Warsaw, Poland , 2020, vol. 170, pp. 634–641.
- Saxen, F.; Werner, P; Handrich, S; Othman, E.; Dinges, L. and Al-Hamadi, A; “Face Attribute Detection with MobileNetV2 and NasNet-Mobile,” 2019, pp. 176–180.
- Handrich, S.; Dinges, L.; Saxen, F. & Al-Hamadi, A, Simultaneous Prediction of Valence / Arousal and Emotion Categories in Real-time,” in IEEE International Conference on Signal and Image Processing Applications (ICSIPA), 2019.
- Dinges, L.; Al-Hamadi, A.; Elzobi, M. & and A. Nürnberger, Automatic Recognition of Common Arabic Handwritten Words based on OCR and N-grams, in International Conference on Image Processing, ICIP 2017, Beijing, China, September 17-20, pp. 3625–3629, 2017
- Dinges, L.; Al-Hamadi, A.; Elzobi, M. & El-etriby, S. Synthetic based Validation of Segmentation of Handwritten Arabic Words, Manuscript Cultures, 2015, 1, 10-18
- Dinges, L.; Al-Hamadi, A. & Elzobi, M. A Locale Group Based Line Segmentation Approach for Non Uniform Skewed and Curved Arabic Handwritings International Conference on Document Analysis and Recognition (ICDAR), 2013, 803-806
- Dinges, L.; Al-Hamadi, A. & Elzobi, M. An Approach for Arabic Handwriting Synthesis based on Active Shape Models International Conference on Document Analysis and Recognition (ICDAR), 2013, 1292-1296
- Dinges, L.; Al-Hamadi, A. & Elzobi, M. An Active Shape Model based approach for Arabic Handwritten Character Recognition The 11th International Conference on Signal Processing ( ICSP'2012), 2012, 1194-1198
- Dinges, L.; Elzobi, M.; Al-Hamadi, A. & Al-Aghbari, Z. Synthizing Handwritten Arabic Text Using Active Shape Models Image Processing and Communications Challenges 3, Springer, 2011, 102, 401-408
- Elzobi, M.; Al-Hamadi, A. & Dinges, L. A Hidden Markov Model-based Approach with an Adaptive Threshold Model for Off-line Arabic Handwriting Recognition International Conference on Document Analysis and Recognition (ICDAR), 2013, 945-949
- Elzobi, M.; Al-Hamadi, A.; Dinges, L & Michaelis., B. A structural feature based segmentation for off-line handwritten Arabic text 5th International Symposium on Image/Video Communication over Fixed and Mobile Networks, ISIVC 2010 . - IEEE, 2010, 1-4
InProceedings (Dinges2023)
Dinges, L.; Fiedler, M.-A.; Al-Hamadi, A.; Abdelrahman, A.; Weimann, J. & Bershadskyy, D.
Foresti, G. L.; Fusiello, A. & Hancock, E. (Eds.)
Uncovering Lies: Deception Detection in a Rolling-Dice Experiment
Image Analysis and Processing -- ICIAP 2023, Springer Nature Switzerland, 2023, 293-303
Abstract: Deception detection is a challenging and interdisciplinary field that has garnered the attention of researchers in psychology, criminology, computer science, and even economics. While automated deception detection presents more obstacles than traditional polygraph tests, it also offers opportunities for novel economic applications. In this study, we propose a novel multimodal approach that combines deep learning with discriminative models to automate deception detection. We tested our approach on two datasets: the Rolling-Dice Experiment, an economically motivated experiment, and a real-life trial dataset for comparison. We utilized video and audio modalities, with video modalities generated through end-to-end learning (CNN). However, for actual deception detection, we employed discriminative approaches due to limited training data in this field. Our results show that the use of multiple modalities and feature selection improves detection results, particularly in the Rolling-Dice Experiment. Furthermore, we observed that due to minimized reactions, deception detection is much more difficult in the Rolling-Dice Experiment than in the high-stake dataset, quantified with an AUC of 0.65 compared to 0.86. Our study highlights the challenges and opportunities of automated deception detection for economic experiments, and our novel multimodal approach shows promise for future research in this field.
Dinges, L.; Fiedler, M.-A.; Al-Hamadi, A.; Abdelrahman, A.; Weimann, J. & Bershadskyy, D.
Foresti, G. L.; Fusiello, A. & Hancock, E. (Eds.)
Uncovering Lies: Deception Detection in a Rolling-Dice Experiment
Image Analysis and Processing -- ICIAP 2023, Springer Nature Switzerland, 2023, 293-303
Abstract: Deception detection is a challenging and interdisciplinary field that has garnered the attention of researchers in psychology, criminology, computer science, and even economics. While automated deception detection presents more obstacles than traditional polygraph tests, it also offers opportunities for novel economic applications. In this study, we propose a novel multimodal approach that combines deep learning with discriminative models to automate deception detection. We tested our approach on two datasets: the Rolling-Dice Experiment, an economically motivated experiment, and a real-life trial dataset for comparison. We utilized video and audio modalities, with video modalities generated through end-to-end learning (CNN). However, for actual deception detection, we employed discriminative approaches due to limited training data in this field. Our results show that the use of multiple modalities and feature selection improves detection results, particularly in the Rolling-Dice Experiment. Furthermore, we observed that due to minimized reactions, deception detection is much more difficult in the Rolling-Dice Experiment than in the high-stake dataset, quantified with an AUC of 0.65 compared to 0.86. Our study highlights the challenges and opportunities of automated deception detection for economic experiments, and our novel multimodal approach shows promise for future research in this field.