Muth, M., Peer, M., Kleber, F., & Sablatnig, R. (2024). Maximizing Data Efficiency of HTR Models by Synthetic Text. In Document Analysis Systems (pp. 295–311). Springer, Cham. https://doi.org/10.1007/978-3-031-70442-0_18
16th IAPR International Workshop on Document Analysis Systems (DAS 2024)
en
Event date:
30-Aug-2024 - 31-Aug-2024
-
Event place:
Athen, Greece
-
Number of Pages:
17
-
Publisher:
Springer, Cham
-
Peer reviewed:
Yes
-
Keywords:
Handwritten Text Recognition; Synthetic Data; Synthetic Text
en
Abstract:
The usability of synthetic handwritten text to improve machine learning models is assessed for the domain of HTR. Synthetic handwritten text is generated using an existing model based on a GAN. The output of this model is then used to train a state-of-the-art HTR model, which is then applied to recognize real datasets. While this results in a CER of 28.3% and a WER of 65.5% for line images of the IAM dataset - more than three times higher than the state-of-the-art result - our experiments show that the amount of real data in a mixed training set can be significantly reduced (70–80%) to achieve comparable CER and WER rates as with real data. Using only 10% of the training data (113 images) from the CVL dataset results in a CER of 54.5% and a WER of 88.8%, pre-training the model with synthetic data results in a CER of 14.6% and a WER of 43.4%.
en
Research Areas:
Visual Computing and Human-Centered Technology: 100%
