SIG4VAM: Synthetic Image Generator for Visual Attention Modeling
This is a code for generating synthetic stimuli
RUN
-Run "stimulusCode.m" to generate stimuli
-Files will be saved to "dataset" and "dataset_blocks"
-both masks and extra files will be generated in such folders
GENERAL FUNCTIONS:
stimulusCode/common_param_values -> defines the experimental setup (image characteristics, object size ...)
stimulusCode/psicometric_param_values -> defines the psychometric values (psi factor, slope, N ...)
stimulusCode/general/ -> it generates any image of the selected type
Please, if you use this code, cite:
https://www.sciencedirect.com/science/article/pii/S0042698918302207 https://openaccess.thecvf.com/content_ICCV_2019/html/Berga_SID4VAM_A_Benchmark_Dataset_With_Synthetic_Images_for_Visual_Attention_ICCV_2019_paper.html
@article{Berga_2019_VisRes,
title = "Psychophysical evaluation of individual low-level feature influences on visual attention",
journal = "Vision Research",
volume = "154",
pages = "60 - 79",
year = "2019",
issn = "0042-6989",
doi = "https://doi.org/10.1016/j.visres.2018.10.006",
url = "http://www.sciencedirect.com/science/article/pii/S0042698918302207",
author = "David Berga and Xosé R. Fdez-Vidal and Xavier Otazu and Víctor Leborán and Xosé M. Pardo"
}
@InProceedings{Berga_2019_ICCV,
author = {Berga, David and Fdez-Vidal, Xose R. and Otazu, Xavier and Pardo, Xose M.},
title = {SID4VAM: A Benchmark Dataset With Synthetic Images for Visual Attention Modeling},
booktitle = {The IEEE International Conference on Computer Vision (ICCV)},
month = {October},
year = {2019}
}
NOTE: The core of the code is based on previous M.W. Spratling's Code
Ref: M. W. Spratling (2012) Predictive coding as a model of the V1 saliency map hypothesis. Neural Networks, 26:7-28.
DOI: 10.1016/j.neunet.2011.10.002
URL=https://nms.kcl.ac.uk/michael.spratling/Code/v1_saliency.zip
SID4VAM: Synthetic Image Dataset for Visual Attention Modeling
Note: You can use this as test set and use the generator (SIG4VAM) with the automatically generated masks to train your model at wish. We use a this benchmark for saliency evaluation.
Images and Masks: https://owncloud.cvc.uab.es/owncloud/index.php/s/BlxkIHdvpN4pIYg
Fixation Data: https://owncloud.cvc.uab.es/owncloud/index.php/s/wfK1SKIjoHEmtzK