English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Multimodal emotion recognition is a challenging task in emotion computing asit is quite difficult to extract discriminative features to identify the subtledifferences in human emotions with abstract concept and multiple expressions.Moreover, how to fully utilize both audio and visual information is still anopen problem. In this paper, we propose a novel multimodal fusion attentionnetwork for audio-visual emotion recognition based on adaptive and multi-levelfactorized bilinear pooling (FBP). First, for the audio stream, a fullyconvolutional network (FCN) equipped with 1-D attention mechanism and localresponse normalization is designed for speech emotion recognition. Next, aglobal FBP (G-FBP) approach is presented to perform audio-visual informationfusion by integrating selfattention based video stream with the proposed audiostream. To improve G-FBP, an adaptive strategy (AG-FBP) to dynamicallycalculate the fusion weight of two modalities is devised based on theemotion-related representation vectors from the attention mechanism ofrespective modalities. Finally, to fully utilize the local emotion information,adaptive and multi-level FBP (AMFBP) is introduced by combining bothglobal-trunk and intratrunk data in one recording on top of AG-FBP. Tested onthe IEMOCAP corpus for speech emotion recognition with only audio stream, thenew FCN method outperforms the state-ofthe-art results with an accuracy of71.40%. Moreover, validated on the AFEW database of EmotiW2019 sub-challengeand the IEMOCAP corpus for audio-visual emotion recognition, the proposedAM-FBP approach achieves the best accuracy of 63.09% and 75.49% respectively onthe test set.

Information Fusion in Attention Networks Using Adaptive and Multi-level Factorized Bilinear Pooling for Audio-visual Emotion Recognition