YOLOv1 to v8: Unveiling Each Variant–A Comprehensive Review of YOLO

Abstract

This paper implements a systematic methodological approach to review the evolution of YOLO variants. Each variant is dissected by examining its internal architectural composition, providing a thorough understanding of its structural components. Subsequently, the review highlights key architectural innovations introduced in each variant, shedding light on the incremental refinements. The review includes benchmarked performance metrics, offering a quantitative measure of each variant's capabilities. The paper further presents the performance of YOLO variants across a diverse range of domains, manifesting their real-world impact. This structured approach ensures a comprehensive examination of YOLOs journey, methodically communicating its internal advancements and benchmarked performance before delving into domain applications. It is envisioned, the incorporation of concepts such as federated learning can introduce a collaborative training paradigm, where YOLO models benefit from training across multiple edge devices, enhancing privacy, adaptability, and generalisation.

Fine-YOLO: A Simplified X-ray Prohibited Object Detection Network Based on Feature Aggregation and Normalized Wasserstein Distance

“…Compared to other YOLO algorithms, YOLOv7 offers the best balance between detection speed and accuracy. By replacing the C3 module in YOLOv5 with the more lightweight C2f module, YOLOv8 [ 26 ] achieves a further improvement in network performance. Despite retaining the SPPF module and the design concept of PAN in the YOLOv5 architecture, YOLOv8 simplifies the network structure by removing the convolutional structure of the sampling stage on the PAN-FPN.…”

HeMoDU: High-Efficiency Multi-Object Detection Algorithm for Unmanned Aerial Vehicles on Urban Roads

“…In recent years, for different detection scenarios and tasks, researchers at home and abroad have proposed various neural network algorithms for object detection, such as DMNet [2], MPFPN [3], NAS-FPN [4], ClusDET [5], and the YOLO [6][7][8] series models. Among them, YOLOv8 is currently the state-of-the-art object detection model, as it can meet real-time requirements while maintaining high precision.…”