---
title: "Category-aware EEG Image Generation Based on Wavelet Transform and Contrast Semantic Loss"
authors:
  - "Enshang Zhang"
  - "Zhicheng Zhang"
  - "Takashi Hanakawa"
doi: "10.24963/ijcai.2025/881"
url: "https://ijcai.org/proceedings/2025/881"
pdf: "https://ijcai.org/proceedings/2025/0881.pdf"
venue: "Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence"
year: 2025
pages: "7922-7930"
track: "Main Track"
keywords:
  - "Computer Vision: CV: Image and video synthesis and generation"
  - "Machine Learning: ML: Attention models"
  - "Multidisciplinary Topics and Applications: MTA: Health and medicine"
  - "Machine Learning: ML: Supervised Learning"
---

# Category-aware EEG Image Generation Based on Wavelet Transform and Contrast Semantic Loss

Enshang Zhang, Zhicheng Zhang, Takashi Hanakawa

*Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence*, pages 7922-7930. DOI: [10.24963/ijcai.2025/881](https://doi.org/10.24963/ijcai.2025/881)

## Abstract

Reconstructing visual stimuli from EEG signals is a crucial step in realizing brain-computer interfaces. In this paper, we propose a transformer-based EEG signal encoder integrating the Discrete Wavelet Transform (DWT) and the gating mechanism. Guided by the feature alignment and category-aware fusion losses, this encoder is used to extract features related to visual stimuli from EEG signals. Subsequently, with the aid of a pre-trained diffusion model, these features are reconstructed into visual stimuli. To verify the effectiveness of the model, we conducted EEG-to-image generation and classification tasks using the THINGS-EEG dataset. To address the limitations of quantitative analysis at the semantic level, we combined WordNet-based classification and semantic similarity metrics to propose a novel semantic-based score, emphasizing the ability of our model to transfer neural activities into visual representations. Experimental results show that our model significantly improves semantic alignment and classification accuracy, which achieves a maximum single-subject accuracy of 43%, outperforming other state-of-the-art methods. The source code is available at https://github.com/zes0v0inn/DWT_EEG_Reconstruction/.