Enregistré dans:
Détails bibliographiques
Auteurs principaux: Jiang, Bowen, Zhuang, Zhijun, Shivakumar, Shreyas S., Taylor, Camillo J.
Format: Preprint
Publié: 2023
Sujets:
Accès en ligne:https://arxiv.org/abs/2311.12889
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866913600302481408
author Jiang, Bowen
Zhuang, Zhijun
Shivakumar, Shreyas S.
Taylor, Camillo J.
author_facet Jiang, Bowen
Zhuang, Zhijun
Shivakumar, Shreyas S.
Taylor, Camillo J.
contents This work introduces an enhanced approach to generating scene graphs by incorporating both a relationship hierarchy and commonsense knowledge. Specifically, we begin by proposing a hierarchical relation head that exploits an informative hierarchical structure. It jointly predicts the relation super-category between object pairs in an image, along with detailed relations under each super-category. Following this, we implement a robust commonsense validation pipeline that harnesses foundation models to critique the results from the scene graph prediction system, removing nonsensical predicates even with a small language-only model. Extensive experiments on Visual Genome and OpenImage V6 datasets demonstrate that the proposed modules can be seamlessly integrated as plug-and-play enhancements to existing scene graph generation algorithms. The results show significant improvements with an extensive set of reasonable predictions beyond dataset annotations. Codes are available at https://github.com/bowen-upenn/scene_graph_commonsense.
format Preprint
id arxiv_https___arxiv_org_abs_2311_12889
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Enhancing Scene Graph Generation with Hierarchical Relationships and Commonsense Knowledge
Jiang, Bowen
Zhuang, Zhijun
Shivakumar, Shreyas S.
Taylor, Camillo J.
Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
This work introduces an enhanced approach to generating scene graphs by incorporating both a relationship hierarchy and commonsense knowledge. Specifically, we begin by proposing a hierarchical relation head that exploits an informative hierarchical structure. It jointly predicts the relation super-category between object pairs in an image, along with detailed relations under each super-category. Following this, we implement a robust commonsense validation pipeline that harnesses foundation models to critique the results from the scene graph prediction system, removing nonsensical predicates even with a small language-only model. Extensive experiments on Visual Genome and OpenImage V6 datasets demonstrate that the proposed modules can be seamlessly integrated as plug-and-play enhancements to existing scene graph generation algorithms. The results show significant improvements with an extensive set of reasonable predictions beyond dataset annotations. Codes are available at https://github.com/bowen-upenn/scene_graph_commonsense.
title Enhancing Scene Graph Generation with Hierarchical Relationships and Commonsense Knowledge
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2311.12889