Distributed Attention for Grounded Image Captioning

TL;DR

This paper introduces a distributed attention mechanism for weakly supervised grounded image captioning, improving the coverage of object regions in generated captions by aggregating information from multiple regions.

Contribution

It proposes a novel distributed attention approach to address partial grounding issues, enhancing the accuracy of region-word alignment in weakly supervised captioning.

Findings

Outperforms state-of-the-art methods in grounded image captioning

Improves coverage of object regions in generated descriptions

Enhances attention accuracy for visually grounded words

Abstract

We study the problem of weakly supervised grounded image captioning. That is, given an image, the goal is to automatically generate a sentence describing the context of the image with each noun word grounded to the corresponding region in the image. This task is challenging due to the lack of explicit fine-grained region word alignments as supervision. Previous weakly supervised methods mainly explore various kinds of regularization schemes to improve attention accuracy. However, their performances are still far from the fully supervised ones. One main issue that has been ignored is that the attention for generating visually groundable words may only focus on the most discriminate parts and can not cover the whole object. To this end, we propose a simple yet effective method to alleviate the issue, termed as partial grounding problem in our paper. Specifically, we design a distributed attention mechanism to enforce the network to aggregate information from multiple spatially different regions with consistent semantics while generating the words. Therefore, the union of the focused region proposals should form a visual region that encloses the object of interest completely. Extensive experiments have demonstrated the superiority of our proposed method compared with the state-of-the-arts.