Constructing Multi-Modal Dialogue Dataset by Replacing Text with Semantically Relevant Images

Abstract

In multi-modal dialogue systems, it is important to allow the use of images as part of a multi-turn conversation. Training such dialogue systems generally requires a large-scale dataset consisting of multi-turn dialogues that involve images, but such datasets rarely exist. In response, this paper proposes a 45k multimodal dialogue dataset created with minimal human intervention. Our method to create such a dataset consists of (1) preparing and pre-processing text dialogue datasets, (2) creating image-mixed dialogues by using a textto-image replacement technique, and (3) employing a contextual-similarity-based filtering step to ensure the contextual coherence of the dataset. To evaluate the validity of our dataset, we devise a simple retrieval model for dialogue sentence prediction tasks. Automatic metrics and human evaluation results on such tasks show that our dataset can be effectively used as training data for multi-modal dialogue systems which require an understanding of images and text in a context-aware manner. Our dataset and generation code is available at https://github.com/shh1574/ multi-modal-dialogue-dataset.

Do as I Demand, Not as I Say: A Dataset for Developing a Reflective Life-Support Robot

“…Existing interactive robots/agents using multimodal features have focused on question-answering from images [14], [15], request analysis [2], and conversations about images [16], [17], [18], [19]. Whether incorporating situation understanding results from multimodal cues significantly improves related tasks has been investigated to determine if we can clearly define things to be recognized for tasks.…”

MMDialog: A Large-scale Multi-turn Dialogue Dataset Towards Multi-modal Open-domain Conversation

“…The topics of human utterances in a dialogue session are often triggered and grounded by these images, which is inconsistent with our daily communications, where the utterances are not always image-related . Secondly, other groups of datasets, such as OpenViDial 1.0/2.0 (Meng et al, 2020; and dialogues collected by Lee et al (2021), are not originated from a real multi-modal conversation scenario. The former directly extracts dialogues and their visual contexts from movies and TV series, and the latter replaces some utterances with retrieved relevant images.…”

“…Then, other groups of works proposed to derive the images from the multi-turn conversations: Meng et al (2020); constructed OpenViDial 1.0/2.0 by directly extracting dialogues and their visual contexts from movies and TV series. Lee et al (2021) also built a multi-modal dialogue dataset by replacing the selected utterances with retrieved relevant images. However, although these corpora were constructed from open-domain conversations with images, they did not originate from a real multi-modal conversation scenario.…”