Altogether: Image Captioning via Re-aligning Alt-text
This addresses the issue of low-quality or opaque captions in image captioning for applications like generative AI and computer vision, though it is incremental as it builds on existing alt-text metadata.
The paper tackled the problem of generating synthetic data for image captioning by re-aligning existing alt-texts to images through multi-round human annotation, resulting in richer captions that improved text-to-image generation and zero-shot image classification tasks.
This paper focuses on creating synthetic data to improve the quality of image captions. Existing works typically have two shortcomings. First, they caption images from scratch, ignoring existing alt-text metadata, and second, lack transparency if the captioners' training data (e.g. GPT) is unknown. In this paper, we study a principled approach Altogether based on the key idea to edit and re-align existing alt-texts associated with the images. To generate training data, we perform human annotation where annotators start with the existing alt-text and re-align it to the image content in multiple rounds, consequently constructing captions with rich visual concepts. This differs from prior work that carries out human annotation as a one-time description task solely based on images and annotator knowledge. We train a captioner on this data that generalizes the process of re-aligning alt-texts at scale. Our results show our Altogether approach leads to richer image captions that also improve text-to-image generation and zero-shot image classification tasks.