StackMix and Blot Augmentations for Handwritten Text Recognition
Alex Shonenkov, Denis Karachev, Maxim Novopoltsev, Mark Potanin, Denis Dimitrov
Code Available — Be the first to reproduce this paper.
ReproduceCode
- github.com/sberbank-ai/StackMix-OCROfficialpytorch★ 48
Abstract
This paper proposes a handwritten text recognition(HTR) system that outperforms current state-of-the-artmethods. The comparison was carried out on three of themost frequently used in HTR task datasets, namely Ben-tham, IAM, and Saint Gall. In addition, the results on tworecently presented datasets, Peter the Greats manuscriptsand HKR Dataset, are provided.The paper describes the architecture of the neural net-work and two ways of increasing the volume of train-ing data: augmentation that simulates strikethrough text(HandWritten Blots) and a new text generation method(StackMix), which proved to be very effective in HTR tasks.StackMix can also be applied to the standalone task of gen-erating handwritten text based on printed text.
Tasks
Benchmark Results
| Dataset | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| Bentham | StackMix+Blots | CER | 1.73 | — | Unverified |
| Digital Peter | StackMix+Blots | CER | 2.5 | — | Unverified |
| HKR | StackMix+Blots | CER | 3.49 | — | Unverified |
| IAM-B | StackMix+Blots | CER | 3.77 | — | Unverified |
| IAM-D | StackMix+Blots | CER | 3.01 | — | Unverified |
| Saint Gall | StackMix+Blots | CER | 3.65 | — | Unverified |