AI Paper Atlas

Research map / Image generation / Text-to-image diffusion

DreamDiffusion: High-Quality EEG-to-Image Generation with Temporal Masked Signal Modeling and CLIP Alignment

Yunpeng Bai, Xintao Wang, Yan-Pei Cao, Yixiao Ge, Chun Yuan, Ying Shan

ECCV 2024 · Image generation › Text-to-image diffusion · approach: Text-driven synthesis and semantic guidance · 21 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add masked signal modeling pretraining on large EEG data (ablation)DreamDiffusion · ImageNet-EEG 50-way top-1accuracy (%): 32.3 -> 45.8Improved
add CLIP image-embedding alignment during fine-tuning (ablation)DreamDiffusion · ImageNet-EEG 50-way top-1accuracy (%): 28.3 -> 45.8Improved
add CLIP supervision without pretraining (ablation)unpretrained EEG encoder · ImageNet-EEG 50-way top-1accuracy (%): 4.2 -> 32.3Improved
use mask ratio 0.75 instead of 0.25 (ablation)MSM pretraining · ImageNet-EEG 50-way top-1accuracy (%): 19.7 -> 45.8Improved
increase mask ratio from 0.75 to 0.85 (ablation)MSM pretraining · ImageNet-EEG 50-way top-1accuracy (%): 45.8 -> 33.4Hurt
increase EEG encoder size to 458M parameters from 297M (ablation)EEG encoder · ImageNet-EEG 50-way top-1accuracy (%): 45.8 -> 38.5Hurt
shrink EEG encoder from 297M to 18.3M parameters (ablation)EEG encoder · ImageNet-EEG 50-way top-1accuracy (%): 45.8 -> 28.7Hurt
fine-tune both EEG encoder and cross-attention heads instead of encoder only (ablation)DreamDiffusion · ImageNet-EEG 50-way top-1accuracy (%): 22.4 -> 45.8Improved
fine-tune both encoder and cross-attention instead of attention only (ablation)DreamDiffusion · ImageNet-EEG 50-way top-1accuracy (%): 20.9 -> 45.8Improved
DreamDiffusion versus Brain2ImageBrain2Image · ImageNet-EEG (Brain2Image categories)FID: 18.76 -> 3.61Improved
DreamDiffusion versus Brain2Image on LPIPS and SSIMBrain2Image · ImageNet-EEG (Brain2Image categories)LPIPS: 0.701 -> 0.644Improved
supervise with category labels instead of CLIPDreamDiffusion · ImageNet-EEG 50-way top-1accuracy (%): 45.8 -> 86.7Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy