Category-Guided Visual Question Generation (Student Abstract)

Visual question generation aims to generate high-quality questions related images. Generating based only on images can better reduce labor costs and thus be easily applied. However, their methods tend similar general that fail ask about the specific content of each image scene. In this paper, we propose a category-guided visual model with multiple categories focus different objects in an image. Specifically, our first selects appropriate category relationships among objects. Then, corresponding selected categories. Experiments conducted TDIUC dataset show proposed outperforms existing models terms diversity quality.