DRISHTIKON: Visual Grounding at Multiple Granularities in Documents

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Kasuba, Badri Vishal, Chaudhuri, Parag, Ramakrishnan, Ganesh
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866918093451689984
author Kasuba, Badri Vishal
Chaudhuri, Parag
Ramakrishnan, Ganesh
author_facet Kasuba, Badri Vishal
Chaudhuri, Parag
Ramakrishnan, Ganesh
contents Visual grounding in text-rich document images is a critical yet underexplored challenge for Document Intelligence and Visual Question Answering (VQA) systems. We present DRISHTIKON, a multi-granular and multi-block visual grounding framework designed to enhance interpretability and trust in VQA for complex, multilingual documents. Our approach integrates multilingual OCR, large language models, and a novel region matching algorithm to localize answer spans at the block, line, word, and point levels. We introduce the Multi-Granular Visual Grounding (MGVG) benchmark, a curated test set of diverse circular notifications from various sectors, each manually annotated with fine-grained, human-verified labels across multiple granularities. Extensive experiments show that our method achieves state-of-the-art grounding accuracy, with line-level granularity providing the best balance between precision and recall. Ablation studies further highlight the benefits of multi-block and multi-line reasoning. Comparative evaluations reveal that leading vision-language models struggle with precise localization, underscoring the effectiveness of our structured, alignment-based approach. Our findings pave the way for more robust and interpretable document understanding systems in real-world, text-centric scenarios with multi-granular grounding support. Code and dataset are made available for future research.
format Preprint
id arxiv_https___arxiv_org_abs_2506_21316
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DRISHTIKON: Visual Grounding at Multiple Granularities in Documents
Kasuba, Badri Vishal
Chaudhuri, Parag
Ramakrishnan, Ganesh
Computer Vision and Pattern Recognition
Visual grounding in text-rich document images is a critical yet underexplored challenge for Document Intelligence and Visual Question Answering (VQA) systems. We present DRISHTIKON, a multi-granular and multi-block visual grounding framework designed to enhance interpretability and trust in VQA for complex, multilingual documents. Our approach integrates multilingual OCR, large language models, and a novel region matching algorithm to localize answer spans at the block, line, word, and point levels. We introduce the Multi-Granular Visual Grounding (MGVG) benchmark, a curated test set of diverse circular notifications from various sectors, each manually annotated with fine-grained, human-verified labels across multiple granularities. Extensive experiments show that our method achieves state-of-the-art grounding accuracy, with line-level granularity providing the best balance between precision and recall. Ablation studies further highlight the benefits of multi-block and multi-line reasoning. Comparative evaluations reveal that leading vision-language models struggle with precise localization, underscoring the effectiveness of our structured, alignment-based approach. Our findings pave the way for more robust and interpretable document understanding systems in real-world, text-centric scenarios with multi-granular grounding support. Code and dataset are made available for future research.
title DRISHTIKON: Visual Grounding at Multiple Granularities in Documents
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2506.21316