OK-VQA

by Kenneth Mario,Mohammed Rastergari,Ali Farhadi,Roozbeth MottaghiResearch Only

OK-VQA

Visual Question Answering (VQA) in its ideal form lets us study reasoning in the joint space of vision and language and serves as a proxy for the AI task of scene understanding. However, most VQA benchmarks to date are focused on questions such as simple counting, visual attributes, and object detection that do not require reasoning or knowledge beyond what is in the image. In this paper, we address the task of knowledge-based visual question answering and provide a benchmark, called OK-VQA, where the image content is not sufficient to answer the questions, encouraging methods that rely on external knowledge resources. Our new dataset includes more than 14,000 questions that require external knowledge to answer. We show that the performance of the state-of-the-art VQA models degrades drastically in this new setting. Our analysis shows that our knowledge-based VQA task is diverse, difficult, and large compared to previous knowledge-based VQA datasets. We hope that this dataset enables researchers to open up new avenues for research in this domain. OK-VQA is a new dataset for visual question answering that requires methods which can draw upon outside knowledge to answer questions. 14,055 open-ended questions 5 ground truth answers per question Manually filtered to ensure all questions require outside knowledge (e.g. from Wikipeida) Reduced questions with most common answers to reduce dataset bias, require external knowledge

Dataset Attributes

Label SVG
TasksVisual Reasoning
Label SVG
CategoriesVisual Question Answering, Questions, Answers, Understanding, External Knowledge, Context
Label SVG
SensorWeb sampling

Class Labels

Vehicles And TransportationsBrands; Companies And ProductsObjects; Material And ClothingSports And RecreationCooking And FoodGeography; History; Language And CulturePeople And Everyday LifePlants And AnimalsScience And TechnologyWeather And Climate