MindEye2 (Not Pretrained) vs. MindEye1

Written by imagerecognition | Published 2025/04/15
Tech Story Tags: mindeye2 | what-is-mindeye2 | mindeye-2-explained | mindeye2-vs-mindeye1 | fmri | image-captioning | one-hour-fmri-decoding | fmri-to-image

TLDRIn this section, we show how MindEye2 outperforms MindEye1 even without pretraining on other subjects.via the TL;DR App

Table of Links

Abstract and 1 Introduction

2 MindEye2 and 2.1 Shared-Subject Functional Alignment

2.2 Backbone, Diffusion Prior, & Submodules

2.3 Image Captioning and 2.4 Fine-tuning Stable Diffusion XL for unCLIP

2.5 Model Inference

3 Results and 3.1 fMRI-to-Image Reconstruction

3.2 Image Captioning

3.3 Image/Brain Retrieval and 3.4 Brain Correlation

3.5 Ablations

4 Related Work

5 Conclusion

6 Acknowledgements and References

A Appendix

A.1 Author Contributions

A.2 Additional Dataset Information

A.3 MindEye2 (not pretrained) vs. MindEye1

A.4 Reconstruction Evaluations Across Varying Amounts of Training Data

A.5 Single-Subject Evaluations

A.6 UnCLIP Evaluation

A.7 OpenCLIP BigG to CLIP L Conversion

A.8 COCO Retrieval

A.9 Reconstruction Evaluations: Additional Information

A.10 Pretraining with Less Subjects

A.11 UMAP Dimensionality Reduction

A.12 ROI-Optimized Stimuli

A.13 Human Preference Experiments

A.3 MindEye2 (not pretrained) vs. MindEye1

Table 6 shows how MindEye2 outperforms MindEye1 even without pretraining on other subjects. Models were trained using the full 40 sessions of training data from subject 1. This suggests that improvements from MindEye1 to MindEye2 are not explained solely from pretraining on other subjects, but that benefits also come from improved model architecture and training procedure.

This paper is available on arxiv under CC BY 4.0 DEED license.

Authors:

(1) Paul S. Scotti, Stability AI and Medical AI Research Center (MedARC);

(2) Mihir Tripathy, Medical AI Research Center (MedARC) and a Core contribution;

(3) Cesar Kadir Torrico Villanueva, Medical AI Research Center (MedARC) and a Core contribution;

(4) Reese Kneeland, University of Minnesota and a Core contribution;

(5) Tong Chen, The University of Sydney and Medical AI Research Center (MedARC);

(6) Ashutosh Narang, Medical AI Research Center (MedARC);

(7) Charan Santhirasegaran, Medical AI Research Center (MedARC);

(8) Jonathan Xu, University of Waterloo and Medical AI Research Center (MedARC);

(9) Thomas Naselaris, University of Minnesota;

(10) Kenneth A. Norman, Princeton Neuroscience Institute;

(11) Tanishq Mathew Abraham, Stability AI and Medical AI Research Center (MedARC).


Written by imagerecognition | Image Recognition
Published by HackerNoon on 2025/04/15