PreSTU: Pre-Training for Scene-Text Understanding
This addresses a key limitation in multimodal AI for applications requiring text-in-image comprehension, though it is incremental as it builds on existing pre-training methods.
The paper tackles the problem of vision-and-language models lacking scene-text understanding by proposing PreSTU, a pre-training recipe with OCR-aware objectives, which improves performance on eight VQA and four image captioning benchmarks.
The ability to recognize and reason about text embedded in visual inputs is often lacking in vision-and-language (V&L) models, perhaps because V&L pre-training methods have often failed to include such an ability in their training objective. In this paper, we propose PreSTU, a novel pre-training recipe dedicated to scene-text understanding (STU). PreSTU introduces OCR-aware pre-training objectives that encourage the model to recognize text from an image and connect it to the rest of the image content. We implement PreSTU using a simple transformer-based encoder-decoder architecture, combined with large-scale image-text datasets with scene text obtained from an off-the-shelf OCR system. We empirically demonstrate the effectiveness of this pre-training approach on eight visual question answering and four image captioning benchmarks.