P-Age: Pexels Dataset for Robust Spatio-Temporal Apparent Age Classification

TL;DR

This paper introduces AgeFormer, a spatio-temporal video-based model utilizing a new dataset, P-Age, to improve age classification accuracy under challenging real-world conditions like occlusion and low resolution.

Contribution

The paper presents a novel two-stream architecture combining TimeSformer and EfficientNet for age estimation from videos, and introduces the P-Age dataset for real-world age classification tasks.

Findings

Outperforms face-based methods in occluded and low-quality scenarios

Effective in diverse challenging video datasets

Demonstrates robustness to occlusions and lighting variations

Abstract

Age estimation is a challenging task that has numerous applications. In this paper, we propose a new direction for age classification that utilizes a video-based model to address challenges such as occlusions, low-resolution, and lighting conditions. To address these challenges, we propose AgeFormer which utilizes spatio-temporal information on the dynamics of the entire body dominating face-based methods for age classification. Our novel two-stream architecture uses TimeSformer and EfficientNet as backbones, to effectively capture both facial and body dynamics information for efficient and accurate age estimation in videos. Furthermore, to fill the gap in predicting age in real-world situations from videos, we construct a video dataset called Pexels Age (P-Age) for age classification. The proposed method achieves superior results compared to existing face-based age estimation methods and is evaluated in situations where the face is highly occluded, blurred, or masked. The method is also cross-tested on a variety of challenging video datasets such as Charades, Smarthome, and Thumos-14.