1. 论文介绍 Patch-Mix Contrastive Learning with Audio Spectrogram Transformer on Respiratory Sound Classification Sangmin Bae1,5∗, June-Woo Kim2,5∗, Won-Yang Cho3,5, Hyerim Baek3,5, Soyoun Son5, B…
论文中,作者提出了一种名为Patch-Mix的对比学习方法,利用Audio Spectrogram Transformer (AST)对呼吸音频进行增强,特别是在处理混合样本的patch特征时,采取了直接的混合策略,确保了模型在ICBHI数据集上的性能提升。原始的mix up方法通过在batch中混合样本,保留部分原始特征并重新计算标签权重,但针对呼吸...
Then, the Patch-Mix feature fusion scheme randomly mixes patches of the enhanced and noisy spectrograms during the Transformer's patch embedding. Furthermore, a novel contrastive learning scheme is introduced to quantify loss and improve model performance, synergizing well with the Transformer model. ...
在COVID19大流行后,非接触医疗护理的需求增加,但医疗数据的获取仍然是一个挑战。Patch Mix对比学习方法通过有效地利用现有数据,增强了模型的学习能力,从而在一定程度上缓解了数据不足的问题。利用Audio Spectrogram Transformer 进行呼吸音频增强:该方法利用AST对呼吸音频进行处理,特别是在处理混合样本的p...