AI Paper Atlas

Research map / Vision-language and multimodal / Cross-modal learning

CineBrain: A Large-Scale Multi-Modal Audiovisual Brain Dataset for Brain-Conditioned Video Generation

Jianxiong Gao, Yichang Liu, Baofeng Yang, Jianfeng Feng, Yanwei Fu

CVPR 2026 · Vision-language and multimodal › Cross-modal learning · approach: Self-supervised audio-visual representation learning

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
fuse fMRI and EEG (CineSync) instead of fMRI only (ablation)CineSync · CineBrain testFVD: 57.47 -> 52.78Improved
fuse fMRI and EEG instead of EEG only (ablation)CineSync · CineBrain testvideo 50-way accuracy: 0.304 -> 0.319Improved
add auditory-cortex ROIs to fMRI input (ablation)CineSync · CineBrain testFVD: 52.78 -> 44.77Improved
add auditory-cortex ROIs to fMRI input (ablation)CineSync · CineBrain testPSNR: 11.99 -> 12.18Improved
add auditory-cortex ROIs to fMRI input (ablation)CineSync · CineBrain testCTC: 0.967 -> 0.953Hurt
encode fMRI and EEG with separate transformers (dual) instead of joint transformer (ablation)Multi-Modal Fusion Encoder · CineBrain testFVD: 128.0 -> 51.53Improved
cross-attention fusion instead of dual transformer fusion (ablation)Multi-Modal Fusion Encoder · CineBrain testFVD: 51.53 -> 110.0Hurt
allocate more tokens to EEG in spatial concatenation (f34-e192 vs f113-e113) (ablation)Spatial concatenation · CineBrain test50-way accuracy: 0.276 -> 0.311Improved
use CineSync-EEG instead of EEG2VideoEEG decoding · CineBrain testFVD: 146.23 -> 53.75Improved
use CineSync-fMRI instead of NeuroClipsfMRI decoding · CineBrain testFVD: 116.36 -> 57.47Improved
add multi-level contrastive alignment (ablation)CineSync · CineBrain testreconstruction qualityImproved

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