Document Type
Article
Journal/Book Title/Conference
Journal of Low Power Electronics and Applications
Volume
16
Issue
1
Publisher
MDPI AG
Publication Date
1-13-2026
Journal Article Version
Version of Record
First Page
1
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Last Page
14
Abstract
In the pursuit of real-time object detection with constrained computational resources, the optimization of neural network architectures is paramount. We introduce novel sparsity induction methods within the YOLOv4-Tiny framework to significantly improve computational efficiency while maintaining high accuracy in pedestrian detection. We present three sparsification approaches: Homogeneous, Progressive, and Layer-Adaptive, each methodically reducing the model’s complexity without compromising its detection capability. Additionally, we refine the model’s output with a memory-efficient sliding window approach and a Bounding Box Sorting Algorithm, ensuring precise Intersection over Union (IoU) calculations. Our results demonstrate a substantial reduction in computational load by zeroing out over 50% of the weights with only a minimal 6% loss in IoU and 0.6% loss in F1-Score.
Recommended Citation
Khan, T.-u.-R.; Roy, S.; Chakraborty, K. Exploring Runtime Sparsification of YOLO Model Weights During Inference. J. Low Power Electron. Appl. 2026, 16, 3. https://doi.org/10.3390/jlpea16010003