Training Affective Computer Vision Models by Crowdsourcing Soft-Target Labels

TL;DR

This paper investigates using crowdsourcing to obtain soft-target labels for emotion classification, demonstrating that filtered crowd labels can effectively capture subjective human interpretations and improve model training on affective datasets.

Contribution

It introduces a method for crowdsourcing soft-target labels for emotion recognition and compares classifiers trained with traditional versus distribution-based labels, highlighting the benefits of capturing label subjectivity.

Findings

Filtered crowd labels align closely with human label distributions.

Classifiers trained with soft labels better reflect human emotion perception.

Crowdsourcing is a feasible approach for acquiring subjective affective labels.

Abstract

Emotion classifiers traditionally predict discrete emotions. However, emotion expressions are often subjective, thus requiring a method to handle subjective labels. We explore the use of crowdsourcing to acquire reliable soft-target labels and evaluate an emotion detection classifier trained with these labels. We center our study on the Child Affective Facial Expression (CAFE) dataset, a gold standard collection of images depicting pediatric facial expressions along with 100 human labels per image. To test the feasibility of crowdsourcing to generate these labels, we used Microworkers to acquire labels for 207 CAFE images. We evaluate both unfiltered workers as well as workers selected through a short crowd filtration process. We then train two versions of a classifiers on soft-target CAFE labels using the original 100 annotations provided with the dataset: (1) a classifier trained with traditional one-hot encoded labels, and (2) a classifier trained with vector labels representing the distribution of CAFE annotator responses. We compare the resulting softmax output distributions of the two classifiers with a 2-sample independent t-test of L1 distances between the classifier's output probability distribution and the distribution of human labels. While agreement with CAFE is weak for unfiltered crowd workers, the filtered crowd agree with the CAFE labels 100% of the time for many emotions. While the F1-score for a one-hot encoded classifier is much higher (94.33% vs. 78.68%) with respect to the ground truth CAFE labels, the output probability vector of the crowd-trained classifier more closely resembles the distribution of human labels (t=3.2827, p=0.0014). Reporting an emotion probability distribution that accounts for the subjectivity of human interpretation. Crowdsourcing, including a sufficient filtering mechanism, is a feasible solution for acquiring soft-target labels.