Towards Khmer Scene Document Layout Detection

TL;DR

This paper introduces a new framework for Khmer scene document layout detection, including a dataset, augmentation tools, and YOLO-based models, addressing the lack of resources and challenges posed by Khmer script complexities.

Contribution

It provides the first comprehensive study with a dedicated dataset, augmentation tools, and baseline models for Khmer scene document layout analysis.

Findings

Created a benchmark dataset for Khmer scene layouts

Developed an augmentation tool for realistic scene document synthesis

Established YOLO-based layout detection baselines with oriented bounding boxes

Abstract

While document layout analysis for Latin scripts has advanced significantly, driven by the advent of large multimodal models (LMMs), progress for the Khmer language remains constrained because of the scarcity of annotated training data. This gap is particularly acute for scene documents, where perspective distortions and complex backgrounds challenge traditional methods. Given the structural complexities of Khmer script, such as diacritics and multi-layer character stacking, existing Latin-based layout analysis models fail to accurately delineate semantic layout units, particularly for dense text regions (e.g., list items). In this paper, we present the first comprehensive study on Khmer scene document layout detection. We contribute a novel framework comprising three key elements: (1) a robust training and benchmarking dataset specifically for Khmer scene layouts; (2) an open-source document augmentation tool capable of synthesizing realistic scene documents to scale training data; and (3) layout detection baselines utilizing YOLO-based architectures with oriented bounding boxes (OBB) to handle geometric distortions. To foster further research in the Khmer document analysis and recognition (DAR) community, we release our models, code, and datasets in this gated repository (in review).