Learning Unsupervised Visual Grounding Through Semantic Self-Supervision

TL;DR

This paper introduces an unsupervised visual grounding framework that leverages concept learning as self-supervision, significantly improving localization accuracy across multiple datasets.

Contribution

It proposes a novel concept learning-based self-supervised approach for unsupervised visual grounding, outperforming existing methods on key benchmarks.

Findings

5.6% improvement on Visual Genome dataset

5.8% improvement on ReferItGame dataset

Comparable performance on Flickr30k

Abstract

Localizing natural language phrases in images is a challenging problem that requires joint understanding of both the textual and visual modalities. In the unsupervised setting, lack of supervisory signals exacerbate this difficulty. In this paper, we propose a novel framework for unsupervised visual grounding which uses concept learning as a proxy task to obtain self-supervision. The simple intuition behind this idea is to encourage the model to localize to regions which can explain some semantic property in the data, in our case, the property being the presence of a concept in a set of images. We present thorough quantitative and qualitative experiments to demonstrate the efficacy of our approach and show a 5.6% improvement over the current state of the art on Visual Genome dataset, a 5.8% improvement on the ReferItGame dataset and comparable to state-of-art performance on the Flickr30k dataset.