DEAP DIVE: Dataset Investigation with Vision transformers for EEG evaluation

2025-10-01

AI Summary

This study addresses the need for low-cost affective computing by investigating emotion recognition performance using sparse-channel EEG signals. To this end, we propose an end-to-end framework integrating Continuous Wavelet Transform (CWT) and Vision Transformer (ViT) on the DEAP dataset: 12-channel EEG time-series signals are first converted into time-frequency representations via CWT; subsequently, ViT directly models spatial-spectral features for quadrant-based emotion classification (valence–arousal space). Experimental results demonstrate that our method achieves 91.57% classification accuracy using only 12 channels—reducing electrode count by 62.5% compared to the standard 32-channel setup—while closely matching full-channel performance. This validates the feasibility of sparse EEG acquisition for affective computing, significantly lowering hardware cost and signal acquisition complexity. The implementation is fully open-sourced to ensure reproducibility.

Technology Category

• Cognitive Modeling & Cognitive Systems: Affective Computing
• Computer Vision: Multi-modal Vision
• Machine Learning: Hardware-aware ML

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Responsible Web: Data and user privacy-enhancing technologies for the Web
• Security and Privacy: Large-scale security measurements

Abstract

Accurately predicting emotions from brain signals has the potential to achieve goals such as improving mental health, human-computer interaction, and affective computing. Emotion prediction through neural signals offers a promising alternative to traditional methods, such as self-assessment and facial expression analysis, which can be subjective or ambiguous. Measurements of the brain activity via electroencephalogram (EEG) provides a more direct and unbiased data source. However, conducting a full EEG is a complex, resource-intensive process, leading to the rise of low-cost EEG devices with simplified measurement capabilities. This work examines how subsets of EEG channels from the DEAP dataset can be used for sufficiently accurate emotion prediction with low-cost EEG devices, rather than fully equipped EEG-measurements. Using Continuous Wavelet Transformation to convert EEG data into scaleograms, we trained a vision transformer (ViT) model for emotion classification. The model achieved over 91,57% accuracy in predicting 4 quadrants (high/low per arousal and valence) with only 12 measuring points (also referred to as channels). Our work shows clearly, that a significant reduction of input channels yields high results compared to state-of-the-art results of 96,9% with 32 channels. Training scripts to reproduce our code can be found here: https://gitlab.kit.edu/kit/aifb/ATKS/public/AutoSMiLeS/DEAP-DIVE.

Problem

• Predicting emotions from reduced EEG channels using vision transformers
• Evaluating low-cost EEG devices for accurate emotion classification
• Achieving high accuracy with minimal EEG measurement points

Innovation

• Using vision transformers for EEG emotion classification
• Converting EEG data into scaleograms via wavelet transformation
• Achieving high accuracy with reduced EEG input channels