Group Affect Prediction Using Multimodal Distributions

TL;DR

This paper presents a CNN-based approach for group emotion prediction from images using multimodal emotion heatmaps, outperforming models trained on raw images on the EmotiW 2017 dataset.

Contribution

The paper introduces a novel CNN training method on emotion heatmaps, demonstrating improved accuracy over raw image models for group emotion recognition.

Findings

Achieved 55.23% validation accuracy on EmotiW 2017 dataset

Outperformed baseline accuracy by 2.44%

Validated effectiveness of emotion heatmaps in group affect prediction

Abstract

We describe our approach towards building an efficient predictive model to detect emotions for a group of people in an image. We have proposed that training a Convolutional Neural Network (CNN) model on the emotion heatmaps extracted from the image, outperforms a CNN model trained entirely on the raw images. The comparison of the models have been done on a recently published dataset of Emotion Recognition in the Wild (EmotiW) challenge, 2017. The proposed method achieved validation accuracy of 55.23% which is 2.44% above the baseline accuracy, provided by the EmotiW organizers.