Interpretable EEG-to-Image Generation with Semantic Prompts
Arshak Rezvani, Ali Akbari, Kosar Sanjar Arani, Maryam Mirian, Emad Arasteh, Martin J. McKeown · 2025
Read Interpretable EEG-to-Image Generation with Semantic Prompts by Arshak Rezvani, Ali Akbari, Kosar Sanjar Arani, Maryam Mirian, Emad Arasteh, Martin J. McKeown (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2507-07157.txt.
Description
Arshak Rezvani and five colleagues present a 2025 paper on turning brain waves into pictures, with words as the middleman. Reading images straight from EEG is hard because scalp signals lack spatial detail. Their system instead matches brain activity to layered captions of a seen image, from plain objects up to abstract themes, written by a language model.
A transformer encoder aligns the EEG with those captions during training. At test time, the matched caption embedding steers a pretrained image generator, producing the picture. The detour through language set state-of-the-art scores on a standard dataset, and the caption levels the brain preferred varied sensibly by image type. Saliency maps showed which scalp regions carried which meanings.
Catalog record
| CREATOR | Arshak Rezvani, Ali Akbari, Kosar Sanjar Arani, Maryam Mirian, Emad Arasteh, Martin J. McKeown |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 13 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2507-07157