Suspected Object Matters: Rethinking Model's Prediction for One-stage Visual Grounding

TL;DR

This paper introduces a novel approach for one-stage visual grounding that focuses on modeling relationships among suspected objects, improving accuracy and efficiency by dynamically identifying and re-evaluating confusing objects during training.

Contribution

The paper proposes the Suspected Object Transformation mechanism (SOT), along with Keyword-Aware Discrimination and Exploration strategies, to enhance one-stage visual grounders by better handling ambiguous objects.

Findings

Significant accuracy improvements on benchmark datasets.

Enhanced model ability to distinguish target objects among confusing candidates.

Effective integration with existing CNN and Transformer-based models.

Abstract

Recently, one-stage visual grounders attract high attention due to their comparable accuracy but significantly higher efficiency than two-stage grounders. However, inter-object relation modeling has not been well studied for one-stage grounders. Inter-object relationship modeling, though important, is not necessarily performed among all objects, as only part of them are related to the text query and may confuse the model. We call these objects suspected objects. However, exploring their relationships in the one-stage paradigm is non-trivial because: First, no object proposals are available as the basis on which to select suspected objects and perform relationship modeling. Second, suspected objects are more confusing than others, as they may share similar semantics, be entangled with certain relationships, etc, and thereby more easily mislead the model prediction. Toward this end, we propose a Suspected Object Transformation mechanism (SOT), which can be seamlessly integrated into existing CNN and Transformer-based one-stage visual grounders to encourage the target object selection among the suspected ones. Suspected objects are dynamically discovered from a learned activation map adapted to the model current discrimination ability during training. Afterward, on top of suspected objects, a Keyword-Aware Discrimination module (KAD) and an Exploration by Random Connection strategy (ERC) are concurrently proposed to help the model rethink its initial prediction. On the one hand, KAD leverages keywords contributing high to suspected object discrimination. On the other hand, ERC allows the model to seek the correct object instead of being trapped in a situation that always exploits the current false prediction. Extensive experiments demonstrate the effectiveness of our proposed method.