Document Image Classification2020en

noisy-bangla-characters

Dataset from Papers With Code

Metrics:accuracy, cer, wer, f1

Current State of the Art

PCGAN-CHAR

Unknown

89.54

accuracy

Showing 2 breakthroughs from Jun 2018 to Aug 2019

Jun 2018

Pixel-level RC

From paper: Pixel-level Reconstruction and Classification for Noisy Handwritten Bangla Characters

77.2

Aug 2019

PCGAN-CHARCurrent SOTA

From paper: PCGAN-CHAR: Progressively Trained Classifier Generative Adversarial Networks for Classification of Noisy Handwritten Bangla Characters

89.5

+16.0%

Total Improvement

16.0%

Time Span

1y 2m

Breakthroughs

Current SOTA

89.5

accuracyPrimary

#	Model	Score	Paper / Code	Date
1	PCGAN-CHAR	89.54	PCGAN-CHAR: Progressively Trained Classifier Generative Adversarial Networks for Classification of Noisy Handwritten Bangla Characters	Aug 2019
2	Pixel-level RC	77.22	Pixel-level Reconstruction and Classification for Noisy Handwritten Bangla Characters	Jun 2018

Aug 2019Models: PCGAN-CHAR

Jun 2018Models: Pixel-level RC