Abstract:Recently, gait-based age and gender recognition have attracted considerable attention in the fields of advertisement marketing and surveillance retrieval due to the unique advantage that gaits can be perceived at a long distance. Intuitively, age and gender can be recognised by observing people's static shape (e.g. different hairstyles between males and females) and dynamic motion (e.g. different walking velocities between the elderly and youth). However, most of the existing gait-based age and gender recognition methods are based on Gait Energy Image (GEI), which loses the capability of explicitly modelling temporal dynamic information and is not robust to the multi-view recognition that inevitably happens in a real application. Therefore, in this study, an Attention-aware Spatio-Temporal Learning (ASTL) framework is proposed, which employs a silhouette sequence as input to learn essential and invariable spatial-temporal gait representations. More specifically, a Multi-Scale Temporal Aggregation (MSTA) module provides an effective scheme for dynamic gait description by exploring and aggregating multi-scale temporal interval information, which is a core supplement to spatial representation. Then, a Multiple Attention Aggregation (MAA) module is designed to help the network focus on the most discriminatory information along temporal, spatial and channel dimensions. Finally, a Multimodal Collaborative Learning (MCL) block gives full play to the advantages of different modal features through a multimodal cooperative learning strategy. The mean absolute error (MAE) for the age estimation and the correct classification rate (CCR) for the gender classification on OU-MVLP achieve 6.68 years and 97%, respectively, demonstrating the superiority of the method. Ablation experiments and visualisation results also prove the effectiveness of the three individual modules in their framework.

Ordinal distribution regression for gait-based age estimation

Gait-Based Age Estimation with Deep Convolutional Neural Network

Gait-based Human Age Estimation Using Age Group-Dependent Manifold Learning and Regression

Gait-Based Human Age Estimation

Ordinal Regression with Multiple Output CNN for Age Estimation

Gait-based age progression/regression: a baseline and performance evaluation by age group classification and cross-age gait identification

Age Estimation Using Expectation of Label Distribution Learning

Learning Ordinal Discriminative Features for Age Estimation

Make the Bag Disappear: Carrying Status-invariant Gait-based Human Age Estimation Using Parallel Generative Adversarial Networks

Gait-based Age Group Classification with Adaptive Graph Neural Network

Learning distance metric regression for facial age estimation

Pgr-Net:A Parallel Network Based On Group And Regression For Age Estimation

Data-Dependent Label Distribution Learning for Age Estimation.

Gait Energy Image-Based Human Attribute Recognition Using Two-Branch Deep Convolutional Neural Network

2D Deep Convolutional Neural Networks for Estimating Stride Length and Velocity in Institutionalized Older Adults

Deep Age Distribution Learning for Apparent Age Estimation

Real-Time Gait-Based Age Estimation and Gender Classification from a Single Image.

Data-dependent label distribution learning for age estimation

Human Age Estimation Based on Locality and Ordinal Information

Partial Least Squares Regression Based Facial Age Estimation

Attention‐aware Spatio‐temporal Learning for Multi‐view Gait‐based Age Estimation and Gender Classification