Visual perception of liquids: insights from deep neural networks

Datasets and analysis code of the following publication:

Van Assen, J.J.R., Nishida, S. & Fleming, R. W. (2020). Visual perception of liquids: insights from deep neural networks. PLOS Computational Biology. DOI: 10.1371/journal.pcbi.1008018

For any questions please contact the first author at mail [at] janjaap [dot] info

Contents:

1. DataAnalysis
- Jupyter Notebook to run the full analysis in R
- For installation details see: https://irkernel.github.io/requirements/

2. FullStimulusSet
- 2 million liquid images with 16 viscosities, 10 scenes, 625 variations, and 20 frames
- Matlab script that merges the images horizontally for network input

3. NeuralActivations
- Matlab files containing the neural activations if you cannot read out the networks

4. TrainedNetworks
- 100 Trained networks referred to in the paper using Matlab and the Deep Learning Toolbox
- One custom layer file “switchLayerAdvanced.m”

5. ValidationSet
- 800 experimental stimuli that were used for validation 16 viscosities, 10 scenes, 5 variations (1,6,11,16,21)
- Matlab script that merges the images horizontally for network input