Towards Models that Can See and Read
원문 초록
Visual Question Answering (VQA) and Image Captioning (CAP), which are among the most popular vision-language tasks, have analogous scene-text versions that require reasoning from the text in the image. Despite their obvious resemblance, the two are treated independently and, as we show, yield task-specific methods that can either see or read, but not both. In this work, we conduct an in-depth analysis of this phenomenon and propose UniTNT, a Unified Text-Non-Text approach, which grants existing multimodal architectures scene-text understanding capabilities. Specifically, we treat scene-text information as an additional modality, fusing it with any pretrained encoder-decoder-based architecture via designated modules. Thorough experiments reveal that UniTNT leads to the first single model that successfully handles both task types. Moreover, we show that scene-text understanding capabilities can boost vision-language models' performance on general VQA and CAP by up to 2.69% and 0.6 CIDEr, respectively.
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
HalLoc: Token-level Localization of Hallucinations for Vision Language Models
Describe Anything Model for Visual Question Answering on Text-rich Images
DrishtiKon: Multi-Granular Visual Grounding for Text-Rich Document Images
HRIBench: Benchmarking Vision-Language Models for Real-Time Human Perception in Human-Robot Interaction
Adapting Lightweight Vision Language Models for Radiological Visual Question Answering
SlotPi: Physics-informed Object-centric Reasoning Models
토론
이 논문과 연결된 포럼 글이 아직 없습니다.