I went to create a model, ResNet and MobileNet , to predict privacy attributes i
I went to create a model, ResNet and MobileNet , to predict privacy attributes in videos. The dataset provided must be used; therefore, alternative datasets are not considered. Preprocessing steps should be applied to the dataset.
Here is the link for the dataset: https://github.com/VITA-Group/PA-HMDB51?tab=readme…
Here is the link for preprocessing the dataset: https://github.com/VITA-Group/Privacy-AdversarialL…
Here is the link for the dataset that will be needed in preprocessing, You will not use this dataset for training or testing the model just for preprocessing. http://serre-lab.clps.brown.edu/wp-content/uploads…
Subsequently, the model will be trained and evaluated in PA-HMDB51 dataest, do not use other dataset and then calculate metrics such as accuracy and F1 score computed to assess its performance on the dataset.
Some testing for the results is needde, as shown in the picture below:
