Dual-Stream EEG Decoding for 3D Visual Perception

The paper presents a dual-stream EEG decoding model designed for 3D shape perception, inspired by the biological vision pathways. It utilizes separate modules for decoding object identity and spatial orientation, employing circular regression for angle prediction and EEG-conditioned multiview diffusion for 3D reconstruction. This work is significant for practitioners as it enhances understanding of neural mechanisms in 3D perception and offers a novel approach to decoding complex visual information from EEG signals.