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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

This paper explores privacy-compliant group-level emotion recognition''in-the-wild'' within the EmotiW Challenge 2023. Group-level emotionrecognition can be useful in many fields including social robotics,conversational agents, e-coaching and learning analytics. This research imposesitself using only global features avoiding individual ones, i.e. all featuresthat can be used to identify or track people in videos (facial landmarks, bodyposes, audio diarization, etc.). The proposed multimodal model is composed of avideo and an audio branches with a cross-attention between modalities. Thevideo branch is based on a fine-tuned ViT architecture. The audio branchextracts Mel-spectrograms and feed them through CNN blocks into a transformerencoder. Our training paradigm includes a generated synthetic dataset toincrease the sensitivity of our model on facial expression within the image ina data-driven way. The extensive experiments show the significance of ourmethodology. Our privacy-compliant proposal performs fairly on the EmotiWchallenge, with 79.24% and 75.13% of accuracy respectively on validation andtest set for the best models. Noticeably, our findings highlight that it ispossible to reach this accuracy level with privacy-compliant features usingonly 5 frames uniformly distributed on the video.

Multimodal Group Emotion Recognition In-the-wild Using Privacy-Compliant Features