Context-VQA: Towards Context-Aware and Purposeful Visual Question Answering
This addresses the need for context-aware VQA to improve accessibility for blind or low-vision users, though it is incremental as it focuses on dataset creation and analysis rather than a new model.
The paper tackled the problem that current VQA models ignore image context, which is crucial for accessibility, by introducing Context-VQA, a dataset with images paired with website contexts, finding that question types vary systematically across contexts (e.g., travel images get 2 times more 'Where?' questions).
Visual question answering (VQA) has the potential to make the Internet more accessible in an interactive way, allowing people who cannot see images to ask questions about them. However, multiple studies have shown that people who are blind or have low-vision prefer image explanations that incorporate the context in which an image appears, yet current VQA datasets focus on images in isolation. We argue that VQA models will not fully succeed at meeting people's needs unless they take context into account. To further motivate and analyze the distinction between different contexts, we introduce Context-VQA, a VQA dataset that pairs images with contexts, specifically types of websites (e.g., a shopping website). We find that the types of questions vary systematically across contexts. For example, images presented in a travel context garner 2 times more "Where?" questions, and images on social media and news garner 2.8 and 1.8 times more "Who?" questions than the average. We also find that context effects are especially important when participants can't see the image. These results demonstrate that context affects the types of questions asked and that VQA models should be context-sensitive to better meet people's needs, especially in accessibility settings.