MICap: A Unified Model for Identity-aware Movie Descriptions

TL;DR

This paper introduces MICap, a unified model for identity-aware movie captioning that can generate captions with character identities or fill-in-the-blanks, improving accuracy and evaluation metrics.

Contribution

MICap is a novel single-stage model that seamlessly switches between id-aware caption generation and fill-in-the-blanks tasks, with a new evaluation metric iSPICE for identity accuracy.

Findings

4.2% improvement in FITB accuracy

1-2% improvement in captioning metrics

Effective unified approach for identity-aware captioning

Abstract

Characters are an important aspect of any storyline and identifying and including them in descriptions is necessary for story understanding. While previous work has largely ignored identity and generated captions with someone (anonymized names), recent work formulates id-aware captioning as a fill-in-the-blanks (FITB) task, where, given a caption with blanks, the goal is to predict person id labels. However, to predict captions with ids, a two-stage approach is required: first predict captions with someone, then fill in identities. In this work, we present a new single stage approach that can seamlessly switch between id-aware caption generation or FITB when given a caption with blanks. Our model, Movie-Identity Captioner (MICap), uses a shared auto-regressive decoder that benefits from training with FITB and full-caption generation objectives, while the encoder can benefit from or disregard captions with blanks as input. Another challenge with id-aware captioning is the lack of a metric to capture subtle differences between person ids. To this end, we introduce iSPICE, a caption evaluation metric that focuses on identity tuples created through intermediate scene graphs. We evaluate MICap on Large-Scale Movie Description Challenge (LSMDC), where we show a 4.2% improvement in FITB accuracy, and a 1-2% bump in classic captioning metrics.