BGF-YOLO: Enhanced YOLOv8 with Multiscale Attentional Feature Fusion for Brain Tumor Detection

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Kang, Ming, Ting, Chee-Ming, Ting, Fung Fung, Phan, Raphaël C. -W.
Formato: Preprint
Publicado: 2023
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866909347049635840
author Kang, Ming
Ting, Chee-Ming
Ting, Fung Fung
Phan, Raphaël C. -W.
author_facet Kang, Ming
Ting, Chee-Ming
Ting, Fung Fung
Phan, Raphaël C. -W.
contents You Only Look Once (YOLO)-based object detectors have shown remarkable accuracy for automated brain tumor detection. In this paper, we develop a novel BGF-YOLO architecture by incorporating Bi-level routing attention, Generalized feature pyramid networks, and Fourth detecting head into YOLOv8. BGF-YOLO contains an attention mechanism to focus more on important features, and feature pyramid networks to enrich feature representation by merging high-level semantic features with spatial details. Furthermore, we investigate the effect of different attention mechanisms and feature fusions, detection head architectures on brain tumor detection accuracy. Experimental results show that BGF-YOLO gives a 4.7% absolute increase of mAP$_{50}$ compared to YOLOv8x, and achieves state-of-the-art on the brain tumor detection dataset Br35H. The code is available at https://github.com/mkang315/BGF-YOLO.
format Preprint
id arxiv_https___arxiv_org_abs_2309_12585
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle BGF-YOLO: Enhanced YOLOv8 with Multiscale Attentional Feature Fusion for Brain Tumor Detection
Kang, Ming
Ting, Chee-Ming
Ting, Fung Fung
Phan, Raphaël C. -W.
Computer Vision and Pattern Recognition
Signal Processing
Applications
68U10 (Primary) 68T10, 68T07, 62P10 (Secondary)
I.4.6; I.5.1; J.3
You Only Look Once (YOLO)-based object detectors have shown remarkable accuracy for automated brain tumor detection. In this paper, we develop a novel BGF-YOLO architecture by incorporating Bi-level routing attention, Generalized feature pyramid networks, and Fourth detecting head into YOLOv8. BGF-YOLO contains an attention mechanism to focus more on important features, and feature pyramid networks to enrich feature representation by merging high-level semantic features with spatial details. Furthermore, we investigate the effect of different attention mechanisms and feature fusions, detection head architectures on brain tumor detection accuracy. Experimental results show that BGF-YOLO gives a 4.7% absolute increase of mAP$_{50}$ compared to YOLOv8x, and achieves state-of-the-art on the brain tumor detection dataset Br35H. The code is available at https://github.com/mkang315/BGF-YOLO.
title BGF-YOLO: Enhanced YOLOv8 with Multiscale Attentional Feature Fusion for Brain Tumor Detection
topic Computer Vision and Pattern Recognition
Signal Processing
Applications
68U10 (Primary) 68T10, 68T07, 62P10 (Secondary)
I.4.6; I.5.1; J.3
url https://arxiv.org/abs/2309.12585