Co-attentional Transformers for Story-Based Video Understanding
This work addresses the challenge of character-centered video story understanding for applications like video question answering, representing an incremental advance in visio-lingual fusion methods.
The paper tackles the problem of story-based video understanding by proposing a co-attentional transformer model to capture long-term dependencies in visual stories, achieving an 8 percentage point overall improvement over the baseline and outperforming the DramaQA challenge winner.
Inspired by recent trends in vision and language learning, we explore applications of attention mechanisms for visio-lingual fusion within an application to story-based video understanding. Like other video-based QA tasks, video story understanding requires agents to grasp complex temporal dependencies. However, as it focuses on the narrative aspect of video it also requires understanding of the interactions between different characters, as well as their actions and their motivations. We propose a novel co-attentional transformer model to better capture long-term dependencies seen in visual stories such as dramas and measure its performance on the video question answering task. We evaluate our approach on the recently introduced DramaQA dataset which features character-centered video story understanding questions. Our model outperforms the baseline model by 8 percentage points overall, at least 4.95 and up to 12.8 percentage points on all difficulty levels and manages to beat the winner of the DramaQA challenge.