Performance Enhancement Leveraging Mask-RCNN on Bengali Document Layout Analysis
This addresses the challenge of machine reading and understanding Bengali documents, which is incremental as it applies an existing method to a new language-specific dataset.
The paper tackled the problem of Document Layout Analysis for Bengali historical documents by training a Mask R-CNN model on the BaDLAD dataset, achieving a dice score of 0.889 through hyperparameter tuning.
Understanding digital documents is like solving a puzzle, especially historical ones. Document Layout Analysis (DLA) helps with this puzzle by dividing documents into sections like paragraphs, images, and tables. This is crucial for machines to read and understand these documents. In the DL Sprint 2.0 competition, we worked on understanding Bangla documents. We used a dataset called BaDLAD with lots of examples. We trained a special model called Mask R-CNN to help with this understanding. We made this model better by step-by-step hyperparameter tuning, and we achieved a good dice score of 0.889. However, not everything went perfectly. We tried using a model trained for English documents, but it didn't fit well with Bangla. This showed us that each language has its own challenges. Our solution for the DL Sprint 2.0 is publicly available at https://www.kaggle.com/competitions/dlsprint2/discussion/432201 along with notebooks, weights, and inference notebook.