Skip to content

Latest commit

 

History

History
20 lines (14 loc) · 1.02 KB

README.md

File metadata and controls

20 lines (14 loc) · 1.02 KB

Continuous and Discrete Expressions detection and visualisation

Emma.gif

This project uses https://github.com/face-analysis/emonet (Paper) to estimate the continuous and discrete emotions in a video. It utilises Facial Alignment Network (FAN) for facial landmark detection, which is used to crop the face image and further fed to EmoNet for expression detection. The results are annotated and on every frame of the video to generate annotated videos with discrete label and changing values of valence and arousal or the continuous values of emotions.

Installation and Usage

Clone the EmoNet repo and install FAN

!git clone https://github.com/face-analysis/emonet
!cp -a /content/emonet/emonet /content/emonet2
!pip install face-alignment

Run Expression detection visualisation.py to generate results.
Input video path can be edited in Line 109 
The 5 or 8 expression model selection can be edited in Line 121