PCGAN-CHAR: Progressively Trained Classifier Generative Adversarial Networks for Classification of Noisy Handwritten Bangla Characters
This addresses classification challenges in noisy handwritten character recognition, particularly for Bangla script, but is incremental as it builds on existing GAN and progressive training methods.
The paper tackles the problem of classifying noisy handwritten characters by training an all-in-one model that classifies without prior denoising, achieving accurate classification on noisy versions of datasets like MNIST and Bangla characters.
Due to the sparsity of features, noise has proven to be a great inhibitor in the classification of handwritten characters. To combat this, most techniques perform denoising of the data before classification. In this paper, we consolidate the approach by training an all-in-one model that is able to classify even noisy characters. For classification, we progressively train a classifier generative adversarial network on the characters from low to high resolution. We show that by learning the features at each resolution independently a trained model is able to accurately classify characters even in the presence of noise. We experimentally demonstrate the effectiveness of our approach by classifying noisy versions of MNIST, handwritten Bangla Numeral, and Basic Character datasets.