ERNIE-ViL: Knowledge Enhanced Vision-Language Representations Through Scene Graph

TL;DR

ERNIE-ViL introduces a knowledge-enhanced method using scene graphs to improve vision-language representations, achieving state-of-the-art results on multiple cross-modal tasks.

Contribution

The paper presents a novel approach that incorporates scene graph prediction tasks into pre-training to enhance detailed semantic alignment across vision and language.

Findings

Achieves state-of-the-art performance on 5 cross-modal tasks.

Ranks first on the VCR leaderboard with a 3.7% improvement.

Effectively models detailed semantic connections between vision and language.

Abstract

We propose a knowledge-enhanced approach, ERNIE-ViL, which incorporates structured knowledge obtained from scene graphs to learn joint representations of vision-language. ERNIE-ViL tries to build the detailed semantic connections (objects, attributes of objects and relationships between objects) across vision and language, which are essential to vision-language cross-modal tasks. Utilizing scene graphs of visual scenes, ERNIE-ViL constructs Scene Graph Prediction tasks, i.e., Object Prediction, Attribute Prediction and Relationship Prediction tasks in the pre-training phase. Specifically, these prediction tasks are implemented by predicting nodes of different types in the scene graph parsed from the sentence. Thus, ERNIE-ViL can learn the joint representations characterizing the alignments of the detailed semantics across vision and language. After pre-training on large scale image-text aligned datasets, we validate the effectiveness of ERNIE-ViL on 5 cross-modal downstream tasks. ERNIE-ViL achieves state-of-the-art performances on all these tasks and ranks the first place on the VCR leaderboard with an absolute improvement of 3.7%.