Underage Detection through a Multi-Task and MultiAge Approach for Screening Minors in Unconstrained Imagery

TL;DR

This paper introduces a multi-task, multi-age deep learning model for accurately detecting minors in unconstrained images, addressing challenges like distribution shift and data imbalance, and demonstrating improved performance on a new comprehensive benchmark.

Contribution

The paper proposes a novel multi-task architecture with dedicated age discrimination heads, combined with data balancing techniques, and evaluates on a new large-scale benchmark for underage detection.

Findings

Reduced mean absolute error from 4.175 to 4.068 years.

Improved under-18 detection F2 score from 0.801 to 0.857.

Achieved nearly 0.99 recall on challenging domain-shifted data.

Abstract

Accurate automatic screening of minors in unconstrained images requires models robust to distribution shift and resilient to the under-representation of children in public datasets. To address these issues, we propose a multi-task architecture with dedicated under/over-age discrimination tasks based on a frozen FaRL vision-language backbone joined with a compact two-layer MLP that shares features across one age-regression head and four binary underage heads (12, 15, 18, and 21 years). This design focuses on the legally critical age range while keeping the backbone frozen. Class imbalance is mitigated through an $\alpha$-reweighted focal loss and age-balanced mini-batch sampling, while an age gap removes ambiguous samples near thresholds. Evaluation is conducted on our new Overall Underage Benchmark (303k cleaned training images, 110k test images), defining both the "ASORES-39k" restricted overall test, which removes the noisiest domains, and the age estimation wild-shifts test "ASWIFT-20k" of 20k-images, stressing extreme poses ($>$45{\deg}), expressions, and low image quality to emulate real-world shifts. Trained on the cleaned overall set with resampling and age gap, our multiage model "F" reduces the mean absolute error on ASORES-39k from 4.175 y (age-only baseline) to 4.068 y and improves under-18 detection from F2 score of 0.801 to 0.857 at 1% false-adult rate. Under the ASWIFT-20k, the same configuration nearly sustains 0.99 recall while F2 rises from 0.742 to 0.833, demonstrating robustness to domain shift.