Do Generative Metrics Predict YOLO Performance? An Evaluation Across Models, Augmentation Ratios, and Dataset Complexity
生成度量是否能预测YOLO性能?在不同模型、增强比例和数据集复杂度上的评估
机构 * School of Mechanical and Mining Engineering, The University of Queensland,Level 4, Mansergh Shaw Building (45), St Lucia, QLD 4072, Australia(昆士兰大学机械与采矿工程学院) ; Swinburne University of Technology, John Street, Hawthorn, VIC 3122, Australia(斯威本科技大学)
专题命中 图像生成评测 :diffusion(abstract);分类 cs.CV
AI总结 本研究评估了生成度量对YOLO性能的预测能力,发现合成增强在不同数据集和场景中对检测mAP有显著提升,但需结合增强量和场景特性进行分析。
Comments 23 pages, 13 figures, includes appendix