papers AI Learner
The Github is limit! Click to go to the new site.

Manifold Mixup improves text recognition with CTC loss

2019-03-11
Bastien Moysset, Ronaldo Messina

Abstract

Modern handwritten text recognition techniques employ deep recurrent neural networks. The use of these techniques is especially efficient when a large amount of annotated data is available for parameter estimation. Data augmentation can be used to enhance the performance of the systems when data is scarce. Manifold Mixup is a modern method of data augmentation that meld two images or the feature maps corresponding to these images and the targets are fused accordingly. We propose to apply the Manifold Mixup to text recognition while adapting it to work with a Connectionist Temporal Classification cost. We show that Manifold Mixup improves text recognition results on various languages and datasets.

Abstract (translated by Google)
URL

https://arxiv.org/abs/1903.04246

PDF

https://arxiv.org/pdf/1903.04246


Similar Posts

Comments