Seeing Straight: Document Orientation Detection for Efficient OCR

TL;DR

This paper introduces a new benchmark and a lightweight, accurate rotation detection method to improve document OCR by ensuring correct document orientation, especially in multilingual and real-world scenarios.

Contribution

The study presents OCR-Rotation-Bench, a new multilingual benchmark, and a fast, robust rotation classification pipeline based on Phi-3.5-Vision, significantly enhancing OCR accuracy.

Findings

Achieved 96% and 92% rotation classification accuracy on benchmark datasets.

Boosted OCR performance by up to 14% with the new rotation module.

Demonstrated effectiveness across multiple languages and real-world conditions.

Abstract

Despite significant advances in document understanding, determining the correct orientation of scanned or photographed documents remains a critical pre-processing step in the real world settings. Accurate rotation correction is essential for enhancing the performance of downstream tasks such as Optical Character Recognition (OCR) where misalignment commonly arises due to user errors, particularly incorrect base orientations of the camera during capture. In this study, we first introduce OCR-Rotation-Bench (ORB), a new benchmark for evaluating OCR robustness to image rotations, comprising (i) ORB-En, built from rotation-transformed structured and free-form English OCR datasets, and (ii) ORB-Indic, a novel multilingual set spanning 11 Indic mid to low-resource languages. We also present a fast, robust and lightweight rotation classification pipeline built on the vision encoder of Phi-3.5-Vision model with dynamic image cropping, fine-tuned specifically for 4-class rotation task in a standalone fashion. Our method achieves near-perfect 96% and 92% accuracy on identifying the rotations respectively on both the datasets. Beyond classification, we demonstrate the critical role of our module in boosting OCR performance: closed-source (up to 14%) and open-weights models (up to 4x) in the simulated real-world setting.