Audio Visual Emotion Recognition with Temporal Alignment and Perception Attention

TL;DR

This paper introduces a novel approach for audio-visual emotion recognition that employs temporal alignment and perception attention mechanisms to improve classification accuracy in video data.

Contribution

It proposes a combined use of soft attention for temporal alignment and perception attention re-weighting within an LSTM-RNN framework, enhancing emotion recognition performance.

Findings

Effective temporal alignment of audio-visual streams demonstrated

Perception attention re-weighting improves emotion classification accuracy

Validated on EmotiW2015 dataset with positive results

Abstract

This paper focuses on two key problems for audio-visual emotion recognition in the video. One is the audio and visual streams temporal alignment for feature level fusion. The other one is locating and re-weighting the perception attentions in the whole audio-visual stream for better recognition. The Long Short Term Memory Recurrent Neural Network (LSTM-RNN) is employed as the main classification architecture. Firstly, soft attention mechanism aligns the audio and visual streams. Secondly, seven emotion embedding vectors, which are corresponding to each classification emotion type, are added to locate the perception attentions. The locating and re-weighting process is also based on the soft attention mechanism. The experiment results on EmotiW2015 dataset and the qualitative analysis show the efficiency of the proposed two techniques.