{"slug": "brain-to-image-generation-reconstructing-visual-stimuli-from-eeg-using-networks", "title": "Brain-to-Image Generation: Reconstructing Visual Stimuli from EEG using Generative Adversarial Networks", "summary": "A single-subject EEG-to-image baseline on the THINGS-EEG2 dataset achieved image recall of 12.83 +/- 0.58%, 39.17 +/- 1.76%, and 58.00 +/- 1.73% at ranks 1, 5, and 10 across three training seeds, against analytical chance levels of 0.5%, 2.5%, and 5.0%, according to the arXiv paper 2609.22282v1. The compact temporal-spatial convolutional encoder maps repetition-averaged EEG (63 by 250) to provided 512-dimensional ViT-B/32 image features, and applying the Subject 01 model to the other nine subjects without adaptation caused a sharp performance drop, exposing subject specificity. Exploratory direct conditional generators trained without external visual weights produced noise-dominated outputs, so the authors conclude the results support above-chance coarse semantic decoding under a closed-set, repetition-averaged protocol but not faithful recovery of stimulus pixels.", "body_md": "arXiv:2609.22282v1 Announce Type: new \nAbstract: Reconstructing visual stimuli from electroencephalography (EEG) is difficult because scalp measurements have high temporal but limited spatial resolution, and paired EEG-image datasets remain small relative to modern generative-model training corpora. We present a reproducible single-subject baseline on THINGS-EEG2 that first tests the more defensible question of whether EEG can retrieve the viewed stimulus in a visual embedding space. A compact temporal-spatial convolutional encoder maps repetition-averaged EEG (63 by 250) to provided 512-dimensional ViT-B/32 image features. Model selection uses a concept-disjoint validation split, and final evaluation uses the official 200-image, 200-concept test gallery. Across three training seeds, the model obtains 12.83 +/- 0.58%, 39.17 +/- 1.76%, and 58.00 +/- 1.73% image recall at 1, 5, and 10 (mean +/- sample standard deviation), compared with analytical chance levels of 0.5%, 2.5%, and 5.0%. A session-balanced ablation shows that averaging more test repetitions generally improves ranking. Applying the Subject 01 model to the other nine subjects without adaptation causes a sharp performance drop, exposing subject specificity. We further report exploratory stress tests of direct conditional generators trained without external visual weights: single-subject and ten-subject variants produce noise-dominated outputs, with early validation improvements reversing after one to four epochs. Finally, we distinguish direct reconstruction from semantic rendering with a pretrained diffusion prior. The results support above-chance coarse semantic decoding under a closed-set, repetition-averaged protocol, but do not support faithful recovery of stimulus pixels.", "url": "https://wpnews.pro/news/brain-to-image-generation-reconstructing-visual-stimuli-from-eeg-using-networks", "canonical_source": "https://arxiv.org/abs/2609.22282", "published_at": "2026-09-22 04:00:00+00:00", "updated_at": "2026-09-22 04:26:36.728594+00:00", "lang": "en", "topics": ["machine-learning", "neural-networks", "computer-vision", "generative-ai", "ai-research"], "entities": ["THINGS-EEG2", "ViT-B/32", "arXiv"], "alternates": {"html": "https://wpnews.pro/news/brain-to-image-generation-reconstructing-visual-stimuli-from-eeg-using-networks", "markdown": "https://wpnews.pro/news/brain-to-image-generation-reconstructing-visual-stimuli-from-eeg-using-networks.md", "text": "https://wpnews.pro/news/brain-to-image-generation-reconstructing-visual-stimuli-from-eeg-using-networks.txt", "jsonld": "https://wpnews.pro/news/brain-to-image-generation-reconstructing-visual-stimuli-from-eeg-using-networks.jsonld"}}