MindEye2: Shared-Subject Models Enable fMRI-To-Image With 1 Hour of Data
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866909224415526912 |
|---|---|
| author | Scotti, Paul S. Tripathy, Mihir Villanueva, Cesar Kadir Torrico Kneeland, Reese Chen, Tong Narang, Ashutosh Santhirasegaran, Charan Xu, Jonathan Naselaris, Thomas Norman, Kenneth A. Abraham, Tanishq Mathew |
| author_facet | Scotti, Paul S. Tripathy, Mihir Villanueva, Cesar Kadir Torrico Kneeland, Reese Chen, Tong Narang, Ashutosh Santhirasegaran, Charan Xu, Jonathan Naselaris, Thomas Norman, Kenneth A. Abraham, Tanishq Mathew |
| contents | Reconstructions of visual perception from brain activity have improved tremendously, but the practical utility of such methods has been limited. This is because such models are trained independently per subject where each subject requires dozens of hours of expensive fMRI training data to attain high-quality results. The present work showcases high-quality reconstructions using only 1 hour of fMRI training data. We pretrain our model across 7 subjects and then fine-tune on minimal data from a new subject. Our novel functional alignment procedure linearly maps all brain data to a shared-subject latent space, followed by a shared non-linear mapping to CLIP image space. We then map from CLIP space to pixel space by fine-tuning Stable Diffusion XL to accept CLIP latents as inputs instead of text. This approach improves out-of-subject generalization with limited training data and also attains state-of-the-art image retrieval and reconstruction metrics compared to single-subject approaches. MindEye2 demonstrates how accurate reconstructions of perception are possible from a single visit to the MRI facility. All code is available on GitHub. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2403_11207 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | MindEye2: Shared-Subject Models Enable fMRI-To-Image With 1 Hour of Data Scotti, Paul S. Tripathy, Mihir Villanueva, Cesar Kadir Torrico Kneeland, Reese Chen, Tong Narang, Ashutosh Santhirasegaran, Charan Xu, Jonathan Naselaris, Thomas Norman, Kenneth A. Abraham, Tanishq Mathew Computer Vision and Pattern Recognition Artificial Intelligence Neurons and Cognition Reconstructions of visual perception from brain activity have improved tremendously, but the practical utility of such methods has been limited. This is because such models are trained independently per subject where each subject requires dozens of hours of expensive fMRI training data to attain high-quality results. The present work showcases high-quality reconstructions using only 1 hour of fMRI training data. We pretrain our model across 7 subjects and then fine-tune on minimal data from a new subject. Our novel functional alignment procedure linearly maps all brain data to a shared-subject latent space, followed by a shared non-linear mapping to CLIP image space. We then map from CLIP space to pixel space by fine-tuning Stable Diffusion XL to accept CLIP latents as inputs instead of text. This approach improves out-of-subject generalization with limited training data and also attains state-of-the-art image retrieval and reconstruction metrics compared to single-subject approaches. MindEye2 demonstrates how accurate reconstructions of perception are possible from a single visit to the MRI facility. All code is available on GitHub. |
| title | MindEye2: Shared-Subject Models Enable fMRI-To-Image With 1 Hour of Data |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Neurons and Cognition |
| url | https://arxiv.org/abs/2403.11207 |