Scores on benchmarks
Model rank shown below is with respect to all public models.| .072 |
average_vision
rank 427
111 benchmarks |
|
| .006 |
neural_vision
rank 508
68 benchmarks |
|
| .025 |
V1
rank 492
31 benchmarks |
|
| .178 |
Coggan2024_fMRI.V1-rdm
v1
rank 38
|
|
|
24 images 
|
||
| .138 |
behavior_vision
rank 274
43 benchmarks |
|
| .098 |
Geirhos2021-error_consistency
[reference]
rank 242
17 benchmarks |
|
| .179 |
Geirhos2021colour-error_consistency
v1
[reference]
rank 214
|
|
|
640 images 
|
||
| .079 |
Geirhos2021contrast-error_consistency
v1
[reference]
rank 249
|
|
|
800 images 
|
||
| .160 |
Geirhos2021cueconflict-error_consistency
v1
[reference]
rank 188
|
|
|
1280 images 
|
||
| .084 |
Geirhos2021edge-error_consistency
v1
[reference]
rank 156
|
|
|
160 images 
|
||
| .077 |
Geirhos2021eidolonI-error_consistency
v1
[reference]
rank 298
|
|
|
800 images 
|
||
| .173 |
Geirhos2021eidolonII-error_consistency
v1
[reference]
rank 242
|
|
|
640 images 
|
||
| .123 |
Geirhos2021eidolonIII-error_consistency
v1
[reference]
rank 261
|
|
|
480 images 
|
||
| .218 |
Geirhos2021falsecolour-error_consistency
v1
[reference]
rank 196
|
|
|
560 images 
|
||
| .048 |
Geirhos2021highpass-error_consistency
v1
[reference]
rank 198
|
|
|
640 images 
|
||
| .090 |
Geirhos2021lowpass-error_consistency
v1
[reference]
rank 224
|
|
|
800 images 
|
||
| .051 |
Geirhos2021phasescrambling-error_consistency
v1
[reference]
rank 241
|
|
|
640 images 
|
||
| .013 |
Geirhos2021powerequalisation-error_consistency
v1
[reference]
rank 304
|
|
|
560 images 
|
||
| .063 |
Geirhos2021rotation-error_consistency
v1
[reference]
rank 234
|
|
|
960 images 
|
||
| .052 |
Geirhos2021silhouette-error_consistency
v1
[reference]
rank 311
|
|
|
160 images 
|
||
| .044 |
Geirhos2021sketch-error_consistency
v1
[reference]
rank 249
|
|
|
800 images 
|
||
| .159 |
Geirhos2021stylized-error_consistency
v1
[reference]
rank 211
|
|
|
800 images 
|
||
| .061 |
Geirhos2021uniformnoise-error_consistency
v1
[reference]
rank 217
|
|
|
800 images 
|
||
| .091 |
Baker2022
rank 186
3 benchmarks |
|
| .274 |
Baker2022fragmented-accuracy_delta
v1
[reference]
rank 143
|
|
|
716 images 
|
||
| .000 |
Baker2022frankenstein-accuracy_delta
v1
[reference]
rank 175
|
|
|
716 images 
|
||
| .000 |
Baker2022inverted-accuracy_delta
v1
[reference]
rank 71
|
|
|
360 images 
|
||
| .146 |
BMD2024
rank 152
4 benchmarks |
|
| .166 |
BMD2024.dotted_1Behavioral-accuracy_distance
v1
rank 101
|
|
|
100 images 
|
||
| .126 |
BMD2024.dotted_2Behavioral-accuracy_distance
v1
rank 132
|
|
|
100 images 
|
||
| .134 |
BMD2024.texture_1Behavioral-accuracy_distance
v1
rank 157
|
|
|
100 images 
|
||
| .157 |
BMD2024.texture_2Behavioral-accuracy_distance
v1
rank 143
|
|
|
100 images 
|
||
| .019 |
Ferguson2024
[reference]
rank 295
14 benchmarks |
|
| .270 |
Ferguson2024gray_hard-value_delta
v1
[reference]
rank 210
|
|
|
2_way_afc task
48 images 
|
||
| .233 |
Hebart2023-match
v1
rank 175
|
|
|
1854 images 
|
||
| .417 |
Maniquet2024
rank 226
2 benchmarks |
|
| .164 |
Maniquet2024-confusion_similarity
v1
[reference]
rank 261
|
|
|
13600 images 
|
||
| .671 |
Maniquet2024-tasks_consistency
v1
[reference]
rank 99
|
|
|
13600 images 
|
||
| .095 |
Coggan2024_behavior-ConditionWiseAccuracySimilarity
v1
rank 205
|
|
|
22560 images 
|
||
| .104 |
engineering_vision
rank 306
25 benchmarks |
|
| .520 |
Geirhos2021-top1
[reference]
rank 182
17 benchmarks |
|
| .948 |
Geirhos2021colour-top1
v1
[reference]
rank 177
|
|
|
640 images 
|
||
| .855 |
Geirhos2021contrast-top1
v1
[reference]
rank 120
|
|
|
800 images 
|
||
| .158 |
Geirhos2021cueconflict-top1
v1
[reference]
rank 263
|
|
|
1280 images 
|
||
| .194 |
Geirhos2021edge-top1
v1
[reference]
rank 230
|
|
|
160 images 
|
||
| .498 |
Geirhos2021eidolonI-top1
v1
[reference]
rank 162
|
|
|
800 images 
|
||
| .511 |
Geirhos2021eidolonII-top1
v1
[reference]
rank 170
|
|
|
640 images 
|
||
| .498 |
Geirhos2021eidolonIII-top1
v1
[reference]
rank 190
|
|
|
480 images 
|
||
| .929 |
Geirhos2021falsecolour-top1
v1
[reference]
rank 159
|
|
|
560 images 
|
||
| .500 |
Geirhos2021highpass-top1
v1
[reference]
rank 88
|
|
|
640 images 
|
||
| .419 |
Geirhos2021lowpass-top1
v1
[reference]
rank 161
|
|
|
800 images 
|
||
| .539 |
Geirhos2021phasescrambling-top1
v1
[reference]
rank 218
|
|
|
640 images 
|
||
| .598 |
Geirhos2021powerequalisation-top1
v1
[reference]
rank 210
|
|
|
560 images 
|
||
| .674 |
Geirhos2021rotation-top1
v1
[reference]
rank 151
|
|
|
960 images 
|
||
| .319 |
Geirhos2021silhouette-top1
v1
[reference]
rank 262
|
|
|
160 images 
|
||
| .488 |
Geirhos2021sketch-top1
v1
[reference]
rank 243
|
|
|
800 images 
|
||
| .329 |
Geirhos2021stylized-top1
v1
[reference]
rank 237
|
|
|
800 images 
|
||
| .381 |
Geirhos2021uniformnoise-top1
v1
[reference]
rank 180
|
|
|
800 images 
|
||
How to use
from brainscore_vision import load_model
model = load_model("resnet50_eMMCR_Vanilla")
model.start_task(...)
model.start_recording(...)
model.look_at(...)
Brain Encoding Response Generator (BERG)
Through the BERG you can easily generate neural responses to images of your choice using any Brain-Score vision model.
For more information on how to use BERG, see the documentation and tutorial.
Benchmarks bibtex
@inproceedings{santurkar2019computer,
title={Computer Vision with a Single (Robust) Classifier},
author={Shibani Santurkar and Dimitris Tsipras and Brandon Tran and Andrew Ilyas and Logan Engstrom and Aleksander Madry},
booktitle={ArXiv preprint arXiv:1906.09453},
year={2019}
}
@article{geirhos2021partial,
title={Partial success in closing the gap between human and machine vision},
author={Geirhos, Robert and Narayanappa, Kantharaju and Mitzkus, Benjamin and Thieringer, Tizian and Bethge, Matthias and Wichmann, Felix A and Brendel, Wieland},
journal={Advances in Neural Information Processing Systems},
volume={34},
year={2021},
url={https://openreview.net/forum?id=QkljT4mrfs}
}
@article{BAKER2022104913,
title = {Deep learning models fail to capture the configural nature of human shape perception},
journal = {iScience},
volume = {25},
number = {9},
pages = {104913},
year = {2022},
issn = {2589-0042},
doi = {https://doi.org/10.1016/j.isci.2022.104913},
url = {https://www.sciencedirect.com/science/article/pii/S2589004222011853},
author = {Nicholas Baker and James H. Elder},
keywords = {Biological sciences, Neuroscience, Sensory neuroscience},
abstract = {Summary
A hallmark of human object perception is sensitivity to the holistic configuration of the local shape features of an object. Deep convolutional neural networks (DCNNs) are currently the dominant models for object recognition processing in the visual cortex, but do they capture this configural sensitivity? To answer this question, we employed a dataset of animal silhouettes and created a variant of this dataset that disrupts the configuration of each object while preserving local features. While human performance was impacted by this manipulation, DCNN performance was not, indicating insensitivity to object configuration. Modifications to training and architecture to make networks more brain-like did not lead to configural processing, and none of the networks were able to accurately predict trial-by-trial human object judgements. We speculate that to match human configural sensitivity, networks must be trained to solve a broader range of object tasks beyond category recognition.}
}
@misc{ferguson_ngo_lee_dicarlo_schrimpf_2024,
title={How Well is Visual Search Asymmetry predicted by a Binary-Choice, Rapid, Accuracy-based Visual-search, Oddball-detection (BRAVO) task?},
url={osf.io/5ba3n},
DOI={10.17605/OSF.IO/5BA3N},
publisher={OSF},
author={Ferguson, Michael E, Jr and Ngo, Jerry and Lee, Michael and DiCarlo, James and Schrimpf, Martin},
year={2024},
month={Jun}
}
@article {Maniquet2024.04.02.587669,
author = {Maniquet, Tim and de Beeck, Hans Op and Costantino, Andrea Ivan},
title = {Recurrent issues with deep neural network models of visual recognition},
elocation-id = {2024.04.02.587669},
year = {2024},
doi = {10.1101/2024.04.02.587669},
publisher = {Cold Spring Harbor Laboratory},
URL = {https://www.biorxiv.org/content/early/2024/04/10/2024.04.02.587669},
eprint = {https://www.biorxiv.org/content/early/2024/04/10/2024.04.02.587669.full.pdf},
journal = {bioRxiv}
}