Dixit: Interactive Visual Storytelling via Term Manipulation
This addresses the problem of making visual storytelling more user-controllable and interpretable for general users, though it is incremental as it builds on existing methods like LSTMs and transformers.
The authors tackled interactive visual storytelling by introducing Dixit, a system that lets users iteratively manipulate terms extracted from photo sequences to generate short stories, achieving interpretable and controllable story generation.
In this paper, we introduce Dixit, an interactive visual storytelling system that the user interacts with iteratively to compose a short story for a photo sequence. The user initiates the process by uploading a sequence of photos. Dixit first extracts text terms from each photo which describe the objects (e.g., boy, bike) or actions (e.g., sleep) in the photo, and then allows the user to add new terms or remove existing terms. Dixit then generates a short story based on these terms. Behind the scenes, Dixit uses an LSTM-based model trained on image caption data and FrameNet to distill terms from each image and utilizes a transformer decoder to compose a context-coherent story. Users change images or terms iteratively with Dixit to create the most ideal story. Dixit also allows users to manually edit and rate stories. The proposed procedure opens up possibilities for interpretable and controllable visual storytelling, allowing users to understand the story formation rationale and to intervene in the generation process.