Paper accepted at ACMMM 2019
We are pleased to announce that a paper from OMRON SINIC X was accepted at ACM Multimedia (ACMMM) 2019!
Jiaxin Ma: “Emotion Recognition Using Multimodal Residual LSTM Network”
Overview
In this study, the task is to recognize emotions (including arousal and valence) of human subjects when they are watching different movie clips. We proposed a long-short-term memory (LSTM) network to solve the task.
Our model is featured by a deep multimodal architecture. Two modalities related to human emotions are encephalography (EEG) and peripheral physiological signals (PPS). The network shares temporal weights across modalities in each LSTM layer to learn the correlation between EEG and PPS. Since emotions are evoked concurrently in both modalities, this architecture is focused to find such temporal related features between modalities.
A publicly available dataset for EEG-based emotion recognition, DEAP, is used for the evaluation of our proposed network. We have got promising results that outperformed state-of-the-art methods.
ACM Multimedia
ACM Multimedia is the premier international conference in the area of multimedia within the field of computer science. Multimedia research focuses on the integration of the multiple perspectives offered by different digital modalities including images, text, video, music, sensor data, spoken audio. Since 1993, ACM Multimedia has been bringing together researchers and practitioners from academia and industry to present innovative research results and discuss recent advancements.
Related links
