Neuro-Symbolic Scene Graph Conditioning for Synthetic Image Dataset Generation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Savazzi, Giacomo, Lomurno, Eugenio, Sbrolli, Cristian, Chiatti, Agnese, Matteucci, Matteo
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913750425010176
author Savazzi, Giacomo
Lomurno, Eugenio
Sbrolli, Cristian
Chiatti, Agnese
Matteucci, Matteo
author_facet Savazzi, Giacomo
Lomurno, Eugenio
Sbrolli, Cristian
Chiatti, Agnese
Matteucci, Matteo
contents As machine learning models increase in scale and complexity, obtaining sufficient training data has become a critical bottleneck due to acquisition costs, privacy constraints, and data scarcity in specialised domains. While synthetic data generation has emerged as a promising alternative, a notable performance gap remains compared to models trained on real data, particularly as task complexity grows. Concurrently, Neuro-Symbolic methods, which combine neural networks' learning strengths with symbolic reasoning's structured representations, have demonstrated significant potential across various cognitive tasks. This paper explores the utility of Neuro-Symbolic conditioning for synthetic image dataset generation, focusing specifically on improving the performance of Scene Graph Generation models. The research investigates whether structured symbolic representations in the form of scene graphs can enhance synthetic data quality through explicit encoding of relational constraints. The results demonstrate that Neuro-Symbolic conditioning yields significant improvements of up to +2.59% in standard Recall metrics and +2.83% in No Graph Constraint Recall metrics when used for dataset augmentation. These findings establish that merging Neuro-Symbolic and generative approaches produces synthetic data with complementary structural information that enhances model performance when combined with real data, providing a novel approach to overcome data scarcity limitations even for complex visual reasoning tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2503_17224
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Neuro-Symbolic Scene Graph Conditioning for Synthetic Image Dataset Generation
Savazzi, Giacomo
Lomurno, Eugenio
Sbrolli, Cristian
Chiatti, Agnese
Matteucci, Matteo
Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
As machine learning models increase in scale and complexity, obtaining sufficient training data has become a critical bottleneck due to acquisition costs, privacy constraints, and data scarcity in specialised domains. While synthetic data generation has emerged as a promising alternative, a notable performance gap remains compared to models trained on real data, particularly as task complexity grows. Concurrently, Neuro-Symbolic methods, which combine neural networks' learning strengths with symbolic reasoning's structured representations, have demonstrated significant potential across various cognitive tasks. This paper explores the utility of Neuro-Symbolic conditioning for synthetic image dataset generation, focusing specifically on improving the performance of Scene Graph Generation models. The research investigates whether structured symbolic representations in the form of scene graphs can enhance synthetic data quality through explicit encoding of relational constraints. The results demonstrate that Neuro-Symbolic conditioning yields significant improvements of up to +2.59% in standard Recall metrics and +2.83% in No Graph Constraint Recall metrics when used for dataset augmentation. These findings establish that merging Neuro-Symbolic and generative approaches produces synthetic data with complementary structural information that enhances model performance when combined with real data, providing a novel approach to overcome data scarcity limitations even for complex visual reasoning tasks.
title Neuro-Symbolic Scene Graph Conditioning for Synthetic Image Dataset Generation
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2503.17224