Diverse Generation while Maintaining Semantic Coordination: A Diffusion-Based Data Augmentation Method for Object Detection

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Nie, Sen, Wang, Zhuo, Wang, Xinxin, He, Kun
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866909280491274240
author Nie, Sen
Wang, Zhuo
Wang, Xinxin
He, Kun
author_facet Nie, Sen
Wang, Zhuo
Wang, Xinxin
He, Kun
contents Recent studies emphasize the crucial role of data augmentation in enhancing the performance of object detection models. However,existing methodologies often struggle to effectively harmonize dataset diversity with semantic coordination.To bridge this gap, we introduce an innovative augmentation technique leveraging pre-trained conditional diffusion models to mediate this balance. Our approach encompasses the development of a Category Affinity Matrix, meticulously designed to enhance dataset diversity, and a Surrounding Region Alignment strategy, which ensures the preservation of semantic coordination in the augmented images. Extensive experimental evaluations confirm the efficacy of our method in enriching dataset diversity while seamlessly maintaining semantic coordination. Our method yields substantial average improvements of +1.4AP, +0.9AP, and +3.4AP over existing alternatives on three distinct object detection models, respectively.
format Preprint
id arxiv_https___arxiv_org_abs_2408_02891
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Diverse Generation while Maintaining Semantic Coordination: A Diffusion-Based Data Augmentation Method for Object Detection
Nie, Sen
Wang, Zhuo
Wang, Xinxin
He, Kun
Computer Vision and Pattern Recognition
Recent studies emphasize the crucial role of data augmentation in enhancing the performance of object detection models. However,existing methodologies often struggle to effectively harmonize dataset diversity with semantic coordination.To bridge this gap, we introduce an innovative augmentation technique leveraging pre-trained conditional diffusion models to mediate this balance. Our approach encompasses the development of a Category Affinity Matrix, meticulously designed to enhance dataset diversity, and a Surrounding Region Alignment strategy, which ensures the preservation of semantic coordination in the augmented images. Extensive experimental evaluations confirm the efficacy of our method in enriching dataset diversity while seamlessly maintaining semantic coordination. Our method yields substantial average improvements of +1.4AP, +0.9AP, and +3.4AP over existing alternatives on three distinct object detection models, respectively.
title Diverse Generation while Maintaining Semantic Coordination: A Diffusion-Based Data Augmentation Method for Object Detection
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2408.02891