Enhancing YOLOv8 for Infrared Object Detection via Learnable Multi-Scale Context and Attention
Signal & Image Processing : An International Journal · 14.07.2026 · 17 (1/2/3) · 01-07
DOI: 10.5121/sipij.2026.17301 ↗Kurz gefasst
Fachartikel in der Zeitschrift Signal & Image Processing : An International Journal, online erschienen am 14.07.2026. Verfasst von Anthony Amankwah.
Abstract (Originalsprache)
Infrared (IR) object detection poses unique challenges due to low texture, weak edges, sensor noise, and a strong dependence on global context. Modern convolutional object detectors, including YOLOv8, are primarily optimized for RGB imagery and often underperform when directly applied to large-scale infrared datasets. In this paper, we present a simple yet effective architectural enhancement to YOLOv8 by (1) replacing the standard Spatial Pyramid Pooling Fast (SPPF) module with a learnable multi-scale variant (SPPFPlus), and (2) integrating the Convolutional Block Attention Module (CBAM) into both the backbone and neck. The proposed approach introduces explicit parallel multi-scale context modeling and adaptive channel–spatial attention, addressing key representational limitations of infrared imagery. Extensive experiments on a large infrared dataset demonstrate an improvement of approximately 15% in detection performance over the YOLOv8 baseline, with notable gains in recall and small-object detection. The modifications are lightweight, modular, and can be seamlessly integrated into existing YOLOv8
Zitieren
APA
Amankwah, A. (2026). Enhancing YOLOv8 for Infrared Object Detection via Learnable Multi-Scale Context and Attention. Signal & Image Processing : An International Journal, 17(1/2/3), 01-07. https://doi.org/10.5121/sipij.2026.17301
Vancouver
Amankwah A. Enhancing YOLOv8 for Infrared Object Detection via Learnable Multi-Scale Context and Attention. Signal & Image Processing : An International Journal. 2026;17(1/2/3):01-07. doi:10.5121/sipij.2026.17301
BibTeX
@article{Amankwah2026,
author = {Amankwah, Anthony},
title = {Enhancing YOLOv8 for Infrared Object Detection via Learnable Multi-Scale Context and Attention},
journal = {Signal & Image Processing : An International Journal},
year = {2026},
volume = {17},
number = {1/2/3},
pages = {01--07},
doi = {10.5121/sipij.2026.17301}
}
EUVIMED