NaviDC-OCR: Navigating Document Parsing Across Digital and Camera-Captured Documents

2026-08-13

AI Summary

This work addresses the limitations of existing document parsing methods, which suffer from cascading failures due to inaccurate layout analysis when handling camera-captured documents with geometric distortions, and the challenges faced by end-to-end models in high-resolution settings—such as redundant generation, hallucination, and insufficient structural reasoning. To overcome these issues, the authors propose NaviDC-OCR, a unified framework that integrates vision-language modeling with geometric awareness. The approach enhances robustness to distortion through deformation-aware learning, employs an adaptive sampling mechanism to better represent complex layouts, and explicitly models mathematical expressions and table structures via a content–structure disentanglement strategy. NaviDC-OCR achieves state-of-the-art performance with comprehensive scores of 96.87, 88.53, and 78.41 on OmniDocBench v1.6, Wild-OmniDocBench, and PureDocBench, respectively, and ranks first in the ICDAR 2026 Sci-ImageMiner Challenge.

Abstract

Document parsing aims to transform unstructured documents into structured and machine-readable representations. Recent advances in Vision-Language Models (VLMs) have significantly advanced document parsing. However, existing approaches still face two major challenges. First, decoupled VLM-based methods heavily rely on accurate layout analysis, where geometric distortions in camera-captured documents can introduce cascading errors. Second, although end-to-end VLM-based methods alleviate the dependence on explicit layout detection, they often suffer from redundant generation, hallucinations, and insufficient structural reasoning in high-resolution scenarios. To address these challenges, we propose NaviDC-OCR, a unified framework for document parsing. NaviDC-OCR introduces deformation-aware learning to incorporate geometric perception into VLMs and proposes an adaptive sampling mechanism for complex layout representation. Furthermore, a content-structure decoupled learning strategy is developed to explicitly model formula grammars and table structures, enabling more effective structured representation learning. Extensive experiments demonstrate that NaviDC-OCR achieves state-of-the-art performance across diverse document parsing benchmarks. It obtains overall scores of 96.87, 88.53 and 78.41 on OmniDocBench v1.6, Wild-OmniDocBench, and PureDocBench, respectively, and ranks first in the ICDAR 2026 Sci-ImageMiner Challenge. These results validate the effectiveness and generalization capability of NaviDC-OCR in complex document parsing scenarios.

Problem

• document parsing
• Vision-Language Models
• geometric distortion
• structured representation
• camera-captured documents

Innovation

• deformation-aware learning
• adaptive sampling
• content-structure decoupled learning
• vision-language models
• document parsing