Supplemental Code for "ImpressionNet :A Multi view Approach to Predict Socio Facial Impressions"

Environment requirement

This code is based on Python 3.9, Keras 2.2 and OpenCV 4.5.

Training

In order to train your model, you need to write your own dataloader. The image transforms used for training is in the 'main.py'. Losses used for training is in the loss file and the usage is in the 'main.py'. More details for training can be found in paper.

Testing

We have provided a model trained on annotated AFLW Database for trustworthiness attribute score prediction for testing. You can download it from [Google Drive] (https://drive.google.com/drive/folders/1WRn8qFqozmPJlH0Vev2K7--V6OL7QUiL?usp=sharing), and test it using 'main.py'.

Update - Paper accepted at ICMLC 2022

Paper Link: https://drive.google.com/file/d/1lj-KOLVXjn4_WAs5_fzauvdCuc0lffk9/view

Acceptance Letter: https://drive.google.com/file/d/1hLSjF34aKzfPyydHUI4bC8QtB0P3kHp1/view

Sample Videos

https://drive.google.com/file/d/1aO90Ei_NnDvHmrzT9xT3ERGmbaP-RRzP/view?usp=sharing

https://drive.google.com/file/d/1t1IvGsZpQQaEL8enA_xV9L-OqQVEWEBx/view?usp=sharing