An Efficient and Explanatory Image and Text Clustering System with Multimodal Autoencoder Architecture
This work addresses the need for efficient and explanatory clustering in multimodal data analysis, specifically for summarizing news videos, though it appears incremental as it extends existing autoencoder and LLM tools.
The authors tackled the problem of summarizing videos into thematic clusters by developing a multimodal autoencoder system that combines CNN and LSTM encodings for image and text data, achieving the ability to produce three to five clusters per video with ten LLM-generated phrases per theme.
We demonstrate the efficiencies and explanatory abilities of extensions to the common tools of Autoencoders and LLM interpreters, in the novel context of comparing different cultural approaches to the same international news event. We develop a new Convolutional-Recurrent Variational Autoencoder (CRVAE) model that extends the modalities of previous CVAE models, by using fully-connected latent layers to embed in parallel the CNN encodings of video frames, together with the LSTM encodings of their related text derived from audio. We incorporate the model within a larger system that includes frame-caption alignment, latent space vector clustering, and a novel LLM-based cluster interpreter. We measure, tune, and apply this system to the task of summarizing a video into three to five thematic clusters, with each theme described by ten LLM-produced phrases. We apply this system to two news topics, COVID-19 and the Winter Olympics, and five other topics are in progress.