Natural Scenes Dataset

computer vision image processing machine learning magnetic resonance imaging neuroimaging neuroscience nifti


Here, we collected and pre-processed a massive, high-quality 7T fMRI dataset that can be used to advance our understanding of how the brain works. A unique feature of this dataset is the massive amount of data available per individual subject. The data were acquired using ultra-high-field fMRI (7T, whole-brain, 1.8-mm resolution, 1.6-s TR). We measured fMRI responses while each of 8 participants viewed 9,000–10,000 distinct, color natural scenes (22,500–30,000 trials) in 30–40 weekly scan sessions over the course of a year. Additional measures were collected including resting-state data, retinotopy, category localizers, anatomical data (T1, T2, diffusion, venogram, angiogram), physiological data (pulse, respiration), eye-tracking data, and additional behavioral assessments outside the scanner. Because of its unprecedented scale and richness, NSD can be used to explore diverse neuroscientific questions with high power at the level of individual subjects. In particular, the number of images sampled in this dataset is sufficiently large that the dataset may be of high interest for computer vision, machine learning, and other data-driven applications.

Update Frequency

Currently in preparation. After publication, this dataset is expected to be mostly static, aside from corrections or additions which will be noted in the NSD Data Manual.


Terms & Conditions for the NSD dataset


To receive access to the NSD Data Manual, please fill out the NSD Data Access Agreement Form.

Managed By

Computational Visual Neuroscience Laboratory at the Center for Magnetic Resonance Research

See all datasets managed by Computational Visual Neuroscience Laboratory]( at the [Center for Magnetic Resonance Research.


Resources on AWS

  • Description
    Structural and functional MRI and other data from the Natural Scenes Dataset. Note that objects are currently listable but are not retrievable -- this will remain the case until publication.
    Resource type
    S3 Bucket
    Amazon Resource Name (ARN)
    AWS Region
    AWS CLI Access (No AWS account required)
    aws s3 ls s3://natural-scenes-dataset/ --no-sign-request

Edit this dataset entry on GitHub