CVPR 2026 | Structure-aware label-free test-time calibration for cross-subject EEG-to-image retrieval.
-
Updated
May 28, 2026 - Python
CVPR 2026 | Structure-aware label-free test-time calibration for cross-subject EEG-to-image retrieval.
Personal repository by Guikang Du for BioFormer, an ICML 2026 Poster on cross-subject biomedical time-series classification via spectral structural alignment. Includes lightweight reproduction, experiments, and extensions.
Official code for "Revisiting Subject-Independent Evaluation for Smartphone-Based Fall and Activity Recognition". This repository quantifies the CV-to-LOSO generalization gap on the MobiAct dataset and introduces Feature-Invariance-Conditioned (FIC) pooling with SAM to improve cross-subject HAR performance.
To associate your repository with the cross-subject-generalization topic, visit your repo's landing page and select "manage topics."