A Survey of Neurosymbolic Visual Reasoning with Scene Graphs and Common Sense Knowledge

Tracking #: 689-1669

Flag : Review Received

Authors: 

M. Jaleed Khan
Filip Ilievski
John G. Breslin
Edward Curry

Responsible editor: 

Md Kamruzzaman Sarker

Submission Type: 

Regular Paper

Full PDF Version: 

Cover Letter: 

Dear Editors, I am submitting our survey paper, “A Survey of Neurosymbolic Visual Reasoning with Scene Graphs and Common Sense Knowledge,” to the Neurosymbolic Artificial Intelligence journal. This paper provides a comprehensive survey of the combination of deep learning, common sense knowledge, and neurosymbolic integration for semantic scene representation and visual reasoning. We emphasize the role of scene graph generation, highlight the infusion of common sense knowledge via heterogeneous knowledge graphs, categorize state-of-the-art neurosymbolic approaches and discuss the downstream reasoning tasks, performance evaluation, challenges and future research directions. Given its relevance to current research trends, we believe our paper will be invaluable to researchers in neurosymbolic AI, computer vision and knowledge graphs. Sincerely, Jaleed Khan

Approve Decision: 

Approved

Tags: 

  • Reviewed

Decision:
Major Revision

Solicited Reviews: