MindEye2 (Not Pretrained) vs. MindEye1

15 Apr 2025

In this section, we show how MindEye2 outperforms MindEye1 even without pretraining on other subjects.

MindEye2: Shared-Subject Models Enable fMRI-To-Image With 1 Hour of Data: Single-Subject Evaluations

15 Apr 2025

In this section, we show more exhaustive evaluation metrics computed for every subject individually using 40 hours and 1 hour of fine-tuning data, respectively.

Reconstruction Evaluations Across Varying Amounts of Training Data: Mindeye2

15 Apr 2025

Here, we present a further analysis of how model performance scales with training data.

MindEye2 Dataset: fMRI Preprocessing and Normalization

15 Apr 2025

Provides further details about the fMRI dataset used in MindEye2, including preprocessing and normalization procedures.

Behind the Scenes: The Making of MindEye2

15 Apr 2025

Details the contributions of each author involved in the development and research of MindEye2.

Seeing With Less: MindEye2's Efficient Approach to Brain-Based Image Decoding

12 Apr 2025

We introduce MindEye2, a modeling approach that outputs reconstructions of seen images from fMRI activity with a similar quality to previous approaches

MindEye2: Novel Functional Alignment for fMRI

12 Apr 2025

Discover MindEye2's innovative functional alignment procedure for cross-subject brain decoding, overcoming limitations of traditional methods

Mindeye2: References You Should Check Out

12 Apr 2025

Here are some Mindeye2 references you should check out.

MindEye2: Shared-Subject Models Enable fMRI-To-Image With 1 Hour of Data

11 Apr 2025

Here, we explain where MindEye2 improvements over MindEye1 come from through ablations.