Locate Then Generate: Bridging Vision and Language with Bounding Box for Scene-Text VQA

TL;DR

This paper introduces a novel 'Locate Then Generate' framework for Scene Text VQA that unifies visual and linguistic semantics via bounding boxes, significantly improving accuracy without pre-training.

Contribution

It proposes a new paradigm that explicitly links visual and linguistic information using bounding boxes, enhancing scene text question answering performance.

Findings

Boosts accuracy by over 6% on TextVQA and ST-VQA datasets.

Unifies visual and linguistic modalities through bounding boxes.

Operates effectively without scene text pre-training.

Abstract

In this paper, we propose a novel multi-modal framework for Scene Text Visual Question Answering (STVQA), which requires models to read scene text in images for question answering. Apart from text or visual objects, which could exist independently, scene text naturally links text and visual modalities together by conveying linguistic semantics while being a visual object in an image simultaneously. Different to conventional STVQA models which take the linguistic semantics and visual semantics in scene text as two separate features, in this paper, we propose a paradigm of "Locate Then Generate" (LTG), which explicitly unifies this two semantics with the spatial bounding box as a bridge connecting them. Specifically, at first, LTG locates the region in an image that may contain the answer words with an answer location module (ALM) consisting of a region proposal network and a language refinement network, both of which can transform to each other with one-to-one mapping via the scene text bounding box. Next, given the answer words selected by ALM, LTG generates a readable answer sequence with an answer generation module (AGM) based on a pre-trained language model. As a benefit of the explicit alignment of the visual and linguistic semantics, even without any scene text based pre-training tasks, LTG can boost the absolute accuracy by +6.06% and +6.92% on the TextVQA dataset and the ST-VQA dataset respectively, compared with a non-pre-training baseline. We further demonstrate that LTG effectively unifies visual and text modalities through the spatial bounding box connection, which is underappreciated in previous methods.