An Analysis of Layer-Freezing Strategies for Enhanced Transfer Learning in YOLO Architectures

Andrzej D. Dobrzycki (), Ana M. Bernardos and José R. Casar
Additional contact information
Andrzej D. Dobrzycki: Information Processing and Telecommunications Center, Universidad Politécnica de Madrid, ETSI Telecomunicación, Av. Complutense, 30, 28040 Madrid, Spain
Ana M. Bernardos: Information Processing and Telecommunications Center, Universidad Politécnica de Madrid, ETSI Telecomunicación, Av. Complutense, 30, 28040 Madrid, Spain
José R. Casar: Information Processing and Telecommunications Center, Universidad Politécnica de Madrid, ETSI Telecomunicación, Av. Complutense, 30, 28040 Madrid, Spain

Mathematics, 2025, vol. 13, issue 15, 1-31

Abstract: The You Only Look Once (YOLO) architecture is crucial for real-time object detection. However, deploying it in resource-constrained environments such as unmanned aerial vehicles (UAVs) requires efficient transfer learning. Although layer freezing is a common technique, the specific impact of various freezing configurations on contemporary YOLOv8 and YOLOv10 architectures remains unexplored, particularly with regard to the interplay between freezing depth, dataset characteristics, and training dynamics. This research addresses this gap by presenting a detailed analysis of layer-freezing strategies. We systematically investigate multiple freezing configurations across YOLOv8 and YOLOv10 variants using four challenging datasets that represent critical infrastructure monitoring. Our methodology integrates a gradient behavior analysis (L2 norm) and visual explanations (Grad-CAM) to provide deeper insights into training dynamics under different freezing strategies. Our results reveal that there is no universal optimal freezing strategy but, rather, one that depends on the properties of the data. For example, freezing the backbone is effective for preserving general-purpose features, while a shallower freeze is better suited to handling extreme class imbalance. These configurations reduce graphics processing unit (GPU) memory consumption by up to 28% compared to full fine-tuning and, in some cases, achieve mean average precision (mAP@50) scores that surpass those of full fine-tuning. Gradient analysis corroborates these findings, showing distinct convergence patterns for moderately frozen models. Ultimately, this work provides empirical findings and practical guidelines for selecting freezing strategies. It offers a practical, evidence-based approach to balanced transfer learning for object detection in scenarios with limited resources.

Keywords: YOLO; layer freezing; transfer learning; fine-tuning; object detection (search for similar items in EconPapers)
JEL-codes: C (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
https://www.mdpi.com/2227-7390/13/15/2539/pdf (application/pdf)
https://www.mdpi.com/2227-7390/13/15/2539/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jmathe:v:13:y:2025:i:15:p:2539-:d:1719689

Access Statistics for this article

Mathematics is currently edited by Ms. Emma He

More articles in Mathematics from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().