eeg-denoising

提出了DenoFlow，一种基于整流流（rectified flow）的SSVEP脑电信号去噪方法，通过回归污染信号与干净信号间直线路径的速度场实现去噪，并在两个公开SSVEP数据集上超越了七个基线模型的信号保真度与解码准确率。

This paper proposes Intelligent Partitioning for Self-supervised Denoising (iPSD), a method enabling unsupervised training of deep EEG denoisers by partitioning noisy segments without requiring clean reference data.