GlobalDoc: A Cross-Modal Vision-Language Framework for Real-World Document Image Retrieval and Classification
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2023
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866929575429144576 |
|---|---|
| author | Bakkali, Souhail Biswas, Sanket Ming, Zuheng Coustaty, Mickaël Rusiñol, Marçal Terrades, Oriol Ramos Lladós, Josep |
| author_facet | Bakkali, Souhail Biswas, Sanket Ming, Zuheng Coustaty, Mickaël Rusiñol, Marçal Terrades, Oriol Ramos Lladós, Josep |
| contents | Visual document understanding (VDU) has rapidly advanced with the development of powerful multi-modal language models. However, these models typically require extensive document pre-training data to learn intermediate representations and often suffer a significant performance drop in real-world online industrial settings. A primary issue is their heavy reliance on OCR engines to extract local positional information within document pages, which limits the models' ability to capture global information and hinders their generalizability, flexibility, and robustness. In this paper, we introduce GlobalDoc, a cross-modal transformer-based architecture pre-trained in a self-supervised manner using three novel pretext objective tasks. GlobalDoc improves the learning of richer semantic concepts by unifying language and visual representations, resulting in more transferable models. For proper evaluation, we also propose two novel document-level downstream VDU tasks, Few-Shot Document Image Classification (DIC) and Content-based Document Image Retrieval (DIR), designed to simulate industrial scenarios more closely. Extensive experimentation has been conducted to demonstrate GlobalDoc's effectiveness in practical settings. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2309_05756 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | GlobalDoc: A Cross-Modal Vision-Language Framework for Real-World Document Image Retrieval and Classification Bakkali, Souhail Biswas, Sanket Ming, Zuheng Coustaty, Mickaël Rusiñol, Marçal Terrades, Oriol Ramos Lladós, Josep Computer Vision and Pattern Recognition Visual document understanding (VDU) has rapidly advanced with the development of powerful multi-modal language models. However, these models typically require extensive document pre-training data to learn intermediate representations and often suffer a significant performance drop in real-world online industrial settings. A primary issue is their heavy reliance on OCR engines to extract local positional information within document pages, which limits the models' ability to capture global information and hinders their generalizability, flexibility, and robustness. In this paper, we introduce GlobalDoc, a cross-modal transformer-based architecture pre-trained in a self-supervised manner using three novel pretext objective tasks. GlobalDoc improves the learning of richer semantic concepts by unifying language and visual representations, resulting in more transferable models. For proper evaluation, we also propose two novel document-level downstream VDU tasks, Few-Shot Document Image Classification (DIC) and Content-based Document Image Retrieval (DIR), designed to simulate industrial scenarios more closely. Extensive experimentation has been conducted to demonstrate GlobalDoc's effectiveness in practical settings. |
| title | GlobalDoc: A Cross-Modal Vision-Language Framework for Real-World Document Image Retrieval and Classification |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2309.05756 |