End-to-End Audio Visual Scene-Aware Dialog using Multimodal Attention-Based Video Features
This work addresses scene-aware dialog for real-world applications like human-computer interaction, but it is incremental as it integrates existing technologies from dialog systems, VQA, and video description.
The paper tackles the problem of enabling dialog systems to understand dynamic visual scenes by introducing a new dataset of 9,000 video-based dialogs and training an end-to-end model that generates responses, showing that multimodal attention-based video features enhance dialog quality.
Dialog systems need to understand dynamic visual scenes in order to have conversations with users about the objects and events around them. Scene-aware dialog systems for real-world applications could be developed by integrating state-of-the-art technologies from multiple research areas, including: end-to-end dialog technologies, which generate system responses using models trained from dialog data; visual question answering (VQA) technologies, which answer questions about images using learned image features; and video description technologies, in which descriptions/captions are generated from videos using multimodal information. We introduce a new dataset of dialogs about videos of human behaviors. Each dialog is a typed conversation that consists of a sequence of 10 question-and-answer(QA) pairs between two Amazon Mechanical Turk (AMT) workers. In total, we collected dialogs on roughly 9,000 videos. Using this new dataset for Audio Visual Scene-aware dialog (AVSD), we trained an end-to-end conversation model that generates responses in a dialog about a video. Our experiments demonstrate that using multimodal features that were developed for multimodal attention-based video description enhances the quality of generated dialog about dynamic scenes (videos). Our dataset, model code and pretrained models will be publicly available for a new Video Scene-Aware Dialog challenge.