Expression Recognition in the Wild Using Sequence Modeling
This work addresses expression recognition in uncontrolled environments for human-computer interaction, but it is incremental as it builds on existing methods with a modest improvement over a baseline.
The paper tackles expression recognition in the wild by proposing a bi-modal approach that fuses audio and visual features with sequence-to-sequence models based on GRU and LSTM, achieving an overall accuracy of 41.5% on the Aff-Wild2 database, which is better than the competition baseline of 37%.
As we exceed upon the procedures for modelling the different aspects of behaviour, expression recognition has become a key field of research in Human Computer Interactions. Expression recognition in the wild is a very interesting problem and is challenging as it involves detailed feature extraction and heavy computation. This paper presents the methodologies and techniques used in our contribution to recognize different expressions i.e., neutral, anger, disgust, fear, happiness, sadness, surprise in ABAW competition on Aff-Wild2 database. Aff-Wild2 database consists of videos in the wild labelled for seven different expressions at frame level. We used a bi-modal approach by fusing audio and visual features and train a sequence-to-sequence model that is based on Gated Recurrent Units (GRU) and Long Short Term Memory (LSTM) network. We show experimental results on validation data. The overall accuracy of the proposed approach is 41.5 \%, which is better than the competition baseline of 37\%.