OCRBench: On the Hidden Mystery of OCR in Large Multimodal Models
This work addresses the need for better evaluation of text-related visual tasks in multimodal models, providing a foundational benchmark for researchers, though it is incremental as it focuses on assessment rather than new methods.
The authors tackled the problem of evaluating OCR capabilities in large multimodal models by creating OCRBench, a comprehensive benchmark with 29 datasets, and found that models like GPT4V and Gemini have strengths and weaknesses in handling multilingual text, handwritten text, non-semantic text, and mathematical expression recognition.
Large models have recently played a dominant role in natural language processing and multimodal vision-language learning. However, their effectiveness in text-related visual tasks remains relatively unexplored. In this paper, we conducted a comprehensive evaluation of Large Multimodal Models, such as GPT4V and Gemini, in various text-related visual tasks including Text Recognition, Scene Text-Centric Visual Question Answering (VQA), Document-Oriented VQA, Key Information Extraction (KIE), and Handwritten Mathematical Expression Recognition (HMER). To facilitate the assessment of Optical Character Recognition (OCR) capabilities in Large Multimodal Models, we propose OCRBench, a comprehensive evaluation benchmark. OCRBench contains 29 datasets, making it the most comprehensive OCR evaluation benchmark available. Furthermore, our study reveals both the strengths and weaknesses of these models, particularly in handling multilingual text, handwritten text, non-semantic text, and mathematical expression recognition. Most importantly, the baseline results presented in this study could provide a foundational framework for the conception and assessment of innovative strategies targeted at enhancing zero-shot multimodal techniques. The evaluation pipeline and benchmark are available at https://github.com/Yuliang-Liu/MultimodalOCR.