Learning to Answer Questions in Dynamic Audio-Visual Scenarios
It addresses the problem of multimodal understanding and spatio-temporal reasoning in dynamic audio-visual scenarios for researchers in AI and computer vision, with incremental contributions in dataset creation and model improvement.
The paper tackles the Audio-Visual Question Answering (AVQA) task by introducing the MUSIC-AVQA dataset with over 45K question-answer pairs and developing a spatio-temporal grounded audio-visual network, which outperforms recent approaches by leveraging multisensory perception.
In this paper, we focus on the Audio-Visual Question Answering (AVQA) task, which aims to answer questions regarding different visual objects, sounds, and their associations in videos. The problem requires comprehensive multimodal understanding and spatio-temporal reasoning over audio-visual scenes. To benchmark this task and facilitate our study, we introduce a large-scale MUSIC-AVQA dataset, which contains more than 45K question-answer pairs covering 33 different question templates spanning over different modalities and question types. We develop several baselines and introduce a spatio-temporal grounded audio-visual network for the AVQA problem. Our results demonstrate that AVQA benefits from multisensory perception and our model outperforms recent A-, V-, and AVQA approaches. We believe that our built dataset has the potential to serve as testbed for evaluating and promoting progress in audio-visual scene understanding and spatio-temporal reasoning. Code and dataset: http://gewu-lab.github.io/MUSIC-AVQA/