YOLOv11 Optimization for Efficient Resource Utilization
This work addresses the problem of efficient object detection for computer vision practitioners by providing incremental optimizations to YOLOv11 for better resource utilization.
The research optimized YOLOv11 by creating size-specific modified versions through pruning and reconfiguring layers, tailored to detect objects of different sizes, and introduced an object classifier for model selection. The proposed models improved computational resource efficiency, maintained accuracy, and in some cases outperformed the original YOLOv11, with reduced model sizes and faster inference times.
The objective of this research is to optimize the eleventh iteration of You Only Look Once (YOLOv11) by developing size-specific modified versions of the architecture. These modifications involve pruning unnecessary layers and reconfiguring the main architecture of YOLOv11. Each proposed version is tailored to detect objects of specific size ranges, from small to large. To ensure proper model selection based on dataset characteristics, we introduced an object classifier program. This program identifies the most suitable modified version for a given dataset. The proposed models were evaluated on various datasets and compared with the original YOLOv11 and YOLOv8 models. The experimental results highlight significant improvements in computational resource efficiency, with the proposed models maintaining the accuracy of the original YOLOv11. In some cases, the modified versions outperformed the original model regarding detection performance. Furthermore, the proposed models demonstrated reduced model sizes and faster inference times. Models weights and the object size classifier can be found in this repository