Bilinear Graph Networks for Visual Question Answering

TL;DR

This paper introduces bilinear graph networks that model relationships between words and objects in visual question answering, enabling complex reasoning and achieving state-of-the-art accuracy on VQA v2.0.

Contribution

The paper proposes a novel bilinear graph network approach that models object and word relationships for improved reasoning in visual question answering.

Findings

Achieves 72.41% accuracy on VQA v2.0 test-std set.

Effectively models complex question reasoning.

Outperforms previous state-of-the-art methods.

Abstract

This paper revisits the bilinear attention networks in the visual question answering task from a graph perspective. The classical bilinear attention networks build a bilinear attention map to extract the joint representation of words in the question and objects in the image but lack fully exploring the relationship between words for complex reasoning. In contrast, we develop bilinear graph networks to model the context of the joint embeddings of words and objects. Two kinds of graphs are investigated, namely image-graph and question-graph. The image-graph transfers features of the detected objects to their related query words, enabling the output nodes to have both semantic and factual information. The question-graph exchanges information between these output nodes from image-graph to amplify the implicit yet important relationship between objects. These two kinds of graphs cooperate with each other, and thus our resulting model can model the relationship and dependency between objects, which leads to the realization of multi-step reasoning. Experimental results on the VQA v2.0 validation dataset demonstrate the ability of our method to handle the complex questions. On the test-std set, our best single model achieves state-of-the-art performance, boosting the overall accuracy to 72.41%.