TY - GEN
T1 - Automatic Gender Classification from Handwritten Images
T2 - 19th International Conference on Computer Analysis of Images and Patterns, CAIP 2021
AU - Rabaev, Irina
AU - Litvak, Marina
AU - Asulin, Sean
AU - Tabibi, Or Haim
N1 - Publisher Copyright:
© 2021, Springer Nature Switzerland AG.
PY - 2021/1/1
Y1 - 2021/1/1
N2 - Using a handwritten sample to automatically classify the writer’s gender is an essential task in a wide range of areas, e.g., psychology, historical documents classification, and forensic analysis. The challenge of gender prediction from offline handwriting can be demonstrated by the relatively low (below 90%) performance of state-of-the-art systems. Despite a high interest within a broad spectrum of research communities, the published works in this area generally concentrate on English and Arabic languages. Most of the existing approaches focus on manual feature selection. In this work, we study an application of deep neural networks for gender classification, where we investigate cross-domain transfer learning with ImageNet pre-training. The study was performed on two datasets, the QUWI dataset, consisting of handwritten documents in English and Arabic, and a new dataset of documents in Hebrew script. We perform extensive experiments, analyze and compare the results obtained with different neural networks. We demonstrate that advanced deep-learning models outperform conventional machine learning approaches that were used in previous studies. We also compare the obtained results against human-level performance and show that the problem is challenging for non-experts.
AB - Using a handwritten sample to automatically classify the writer’s gender is an essential task in a wide range of areas, e.g., psychology, historical documents classification, and forensic analysis. The challenge of gender prediction from offline handwriting can be demonstrated by the relatively low (below 90%) performance of state-of-the-art systems. Despite a high interest within a broad spectrum of research communities, the published works in this area generally concentrate on English and Arabic languages. Most of the existing approaches focus on manual feature selection. In this work, we study an application of deep neural networks for gender classification, where we investigate cross-domain transfer learning with ImageNet pre-training. The study was performed on two datasets, the QUWI dataset, consisting of handwritten documents in English and Arabic, and a new dataset of documents in Hebrew script. We perform extensive experiments, analyze and compare the results obtained with different neural networks. We demonstrate that advanced deep-learning models outperform conventional machine learning approaches that were used in previous studies. We also compare the obtained results against human-level performance and show that the problem is challenging for non-experts.
KW - Deep neural network
KW - Gender classification
KW - Offline handwriting analysis
KW - Transfer learning
UR - http://www.scopus.com/inward/record.url?scp=85119506569&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-89131-2_30
DO - 10.1007/978-3-030-89131-2_30
M3 - Conference contribution
AN - SCOPUS:85119506569
SN - 9783030891305
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 329
EP - 339
BT - Computer Analysis of Images and Patterns - 19th International Conference, CAIP 2021, Proceedings
A2 - Tsapatsoulis, Nicolas
A2 - Panayides, Andreas
A2 - Theocharides, Theo
A2 - Lanitis, Andreas
A2 - Lanitis, Andreas
A2 - Pattichis, Constantinos
A2 - Pattichis, Constantinos
A2 - Vento, Mario
PB - Springer Science and Business Media Deutschland GmbH
Y2 - 28 September 2021 through 30 September 2021
ER -