VQA-MHUG
ImagesTextsIntroduced 2021-09-27
VQA-MHUG is a 49-participant dataset of multimodal human gaze on both images and questions during visual question answering (VQA) collected using a high-speed eye tracker.
VQA-MHUG is a 49-participant dataset of multimodal human gaze on both images and questions during visual question answering (VQA) collected using a high-speed eye tracker.