Fast, Diverse and Accurate Image Captioning Guided by Part-Of-Speech

TLDR

Image captioning is inherently ambiguous, and while beam search is the standard approach for generating multiple captions, it is computationally costly and tends to produce generic results; alternative VAE and GAN methods offer diversity but at the expense of accuracy. The study aims to generate diverse, accurate captions by first predicting a meaningful image summary and then using it to guide caption generation. The authors employ part‑of‑speech tags as concise image summaries to steer the caption generation process. The approach yields highly accurate, diverse captions, outperforms beam search in speed, and achieves superior diversity as measured by novel sentences, distinct n‑grams, and mBleu‑4 scores.

Abstract

Image captioning is an ambiguous problem, with many suitable captions for an image. To address ambiguity, beam search is the de facto method for sampling multiple captions. However, beam search is computationally expensive and known to produce generic captions. To address this concern, some variational auto-encoder (VAE) and generative adversarial net (GAN) based methods have been proposed. Though diverse, GAN and VAE are less accurate. In this paper, we first predict a meaningful summary of the image, then generate the caption based on that summary. We use part-of-speech as summaries, since our summary should drive caption generation. We achieve the trifecta: (1) High accuracy for the diverse captions as evaluated by standard captioning metrics and user studies; (2) Faster computation of diverse captions compared to beam search and diverse beam search; and (3) High diversity as evaluated by counting novel sentences, distinct n-grams and mutual overlap (i.e., mBleu-4) scores.