Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | https://arxiv.org/abs/2404.05290 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866911545379782656 |
|---|---|
| author | Biscione, Valerio Montero, Milton L. Dujmovic, Marin Malhotra, Gaurav Yin, Dong Puebla, Guillermo Adolfi, Federico Heaton, Rachel F. Hummel, John E. Evans, Benjamin D. Habashy, Karim Bowers, Jeffrey S. |
| author_facet | Biscione, Valerio Montero, Milton L. Dujmovic, Marin Malhotra, Gaurav Yin, Dong Puebla, Guillermo Adolfi, Federico Heaton, Rachel F. Hummel, John E. Evans, Benjamin D. Habashy, Karim Bowers, Jeffrey S. |
| contents | Multiple benchmarks have been developed to assess the alignment between deep neural networks (DNNs) and human vision. In almost all cases these benchmarks are observational in the sense they are composed of behavioural and brain responses to naturalistic images that have not been manipulated to test hypotheses regarding how DNNs or humans perceive and identify objects. Here we introduce the toolbox \textit{MindSet: Vision}, consisting of a collection of image datasets and related scripts designed to test DNNs on 30 psychological findings. In all experimental conditions, the stimuli are systematically manipulated to test specific hypotheses regarding human visual perception and object recognition. In addition to providing pre-generated datasets of images, we provide code to regenerate these datasets, offering many configurable parameters which greatly extend the dataset versatility for different research contexts, and code to facilitate the testing of DNNs on these image datasets using three different methods (similarity judgments, out-of-distribution classification, and decoder method), accessible via https://github.com/MindSetVision/MindSetVision. To illustrate the challenges these datasets pose for developing better DNN models of human vision, we test several models on range of datasets included in the toolbox. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_05290 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | MindSet: Vision. A toolbox for testing DNNs on key psychological experiments Biscione, Valerio Montero, Milton L. Dujmovic, Marin Malhotra, Gaurav Yin, Dong Puebla, Guillermo Adolfi, Federico Heaton, Rachel F. Hummel, John E. Evans, Benjamin D. Habashy, Karim Bowers, Jeffrey S. Computer Vision and Pattern Recognition Artificial Intelligence Multiple benchmarks have been developed to assess the alignment between deep neural networks (DNNs) and human vision. In almost all cases these benchmarks are observational in the sense they are composed of behavioural and brain responses to naturalistic images that have not been manipulated to test hypotheses regarding how DNNs or humans perceive and identify objects. Here we introduce the toolbox \textit{MindSet: Vision}, consisting of a collection of image datasets and related scripts designed to test DNNs on 30 psychological findings. In all experimental conditions, the stimuli are systematically manipulated to test specific hypotheses regarding human visual perception and object recognition. In addition to providing pre-generated datasets of images, we provide code to regenerate these datasets, offering many configurable parameters which greatly extend the dataset versatility for different research contexts, and code to facilitate the testing of DNNs on these image datasets using three different methods (similarity judgments, out-of-distribution classification, and decoder method), accessible via https://github.com/MindSetVision/MindSetVision. To illustrate the challenges these datasets pose for developing better DNN models of human vision, we test several models on range of datasets included in the toolbox. |
| title | MindSet: Vision. A toolbox for testing DNNs on key psychological experiments |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2404.05290 |