Researchers have developed GSA-YOLO, a new lightweight framework designed for real-time X-ray security inspection. This model, based on YOLOv8n, incorporates structured sparsity and adaptive knowledge distillation to improve detection accuracy and inference speed. GSA-YOLO integrates Group Lasso, Sparse Structure Selection, and an Adaptive Knowledge Distillation mechanism to enhance feature representation and reduce model size. Evaluations on the HiXray and PIDray datasets show GSA-YOLO achieves a leading inference speed of 189.62 FPS with reduced computational cost, alongside improved mAP50:95 scores compared to the baseline. AI
Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →
IMPACT This new framework offers improved speed and accuracy for X-ray security inspections, potentially enhancing threat detection capabilities.
RANK_REASON The cluster contains a research paper detailing a new model and its performance on specific datasets. [lever_c_demoted from research: ic=1 ai=1.0]
