- Title: EEG2Video: Towards Decoding Dynamic Visual Perception from EEG Signals

• Ref: https://openreview.net/forum?id=RfsfRn9OFd
• Summary: Decoding the information from brain activity can enhance the understanding of the brain’s visual processing system. However, previous studies predominantly focus on reconstructing static visual stimuli. In this paper, researchers explore decoding dynamic visual perception from EEG to achieve the following objectives:

1. Constructing a large dataset of EEG signals from 20 subjects while watching 1,400 dynamic video clips representing 40 concepts (EEG-Video Pairs).
2. Annotating each video clip to explore the potential of decoding specific meta-information from EEG (e.g., color, dynamics, human or not, etc.).
3. Proposing a novel baseline, EEG2Video, for video reconstruction from EEG signals using a Seq2Seq architecture.