ComiCap: A VLMs pipeline for dense captioning of Comic Panels
This work addresses the need for comprehensive storyline understanding in comics by enabling automated dense captioning, which is incremental as it builds on existing VLMs and benchmarks.
The authors tackled the problem of generating dense, grounded captions for comic panels by proposing a pipeline using Vision-Language Models (VLMs), which produced captions quantitatively and qualitatively superior to specifically trained models without additional training, and they annotated over 2 million panels across 13,000 books.
The comic domain is rapidly advancing with the development of single- and multi-page analysis and synthesis models. Recent benchmarks and datasets have been introduced to support and assess models' capabilities in tasks such as detection (panels, characters, text), linking (character re-identification and speaker identification), and analysis of comic elements (e.g., dialog transcription). However, to provide a comprehensive understanding of the storyline, a model must not only extract elements but also understand their relationships and generate highly informative captions. In this work, we propose a pipeline that leverages Vision-Language Models (VLMs) to obtain dense, grounded captions. To construct our pipeline, we introduce an attribute-retaining metric that assesses whether all important attributes are identified in the caption. Additionally, we created a densely annotated test set to fairly evaluate open-source VLMs and select the best captioning model according to our metric. Our pipeline generates dense captions with bounding boxes that are quantitatively and qualitatively superior to those produced by specifically trained models, without requiring any additional training. Using this pipeline, we annotated over 2 million panels across 13,000 books, which will be available on the project page https://github.com/emanuelevivoli/ComiCap.