SATTC: Structure-Aware Label-Free Test-Time Calibration for Cross-Subject EEG-to-Image Retrieval
SATTC: 结构感知的无标签测试时校准用于跨主体EEG到图像检索
机构 * Yunnan University(云南大学)
专题命中 EEG解码 :EEG(title,title_cn);neural decoding(abstract)
AI总结 本文提出SATTC,一种无标签测试时校准方法,通过结合几何专家和结构专家,提升跨主体EEG到图像检索的准确性与稳定性。
Comments Accepted to CVPR 2026. Official code: https://github.com/QunjieHuang/SATTC-CVPR2026. This version corrects the dataset naming in the paper text and abstract: the experiments were conducted on THINGS-EEG2, whereas the previous version referred to it as THINGS-EEG in parts of the paper. Citations were already correct. Results are unchanged