Overcoming Occlusions in the Wild: A Multi-Task Age Head Approach to Age Estimation

2025-06-16

AI Summary

Addressing the challenge of age estimation for occluded faces in unconstrained real-world scenarios, this paper proposes a robust multi-task framework integrating generative adversarial networks (GANs) and Transformers. Our key contributions are: (1) a novel SN-Patch GAN-based de-occlusion module that effectively reconstructs occluded facial regions; (2) an ARCM-Swin hybrid feature enhancement architecture that jointly models local texture details and global semantic context; and (3) a Multi-Task Age Head (MTAH) that unifies age regression with probabilistic age distribution learning, regularized by KL divergence. Extensive experiments demonstrate state-of-the-art performance: mean absolute error (MAE) of 3.00, 4.54, and 2.53 years on FG-NET, UTKFace, and MORPH benchmarks, respectively—surpassing prior methods. These results validate the proposed framework’s superior robustness and accuracy under complex, realistic occlusion conditions.

Technology Category

• Computer Vision: Generative Adversarial Networks (GANs) for Vision
• Machine Learning: Deep Generative Models & Autoencoders
• Multiagent Systems: Adversarial Agents

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems

Abstract

Facial age estimation has achieved considerable success under controlled conditions. However, in unconstrained real-world scenarios, which are often referred to as 'in the wild', age estimation remains challenging, especially when faces are partially occluded, which may obscure their visibility. To address this limitation, we propose a new approach integrating generative adversarial networks (GANs) and transformer architectures to enable robust age estimation from occluded faces. We employ an SN-Patch GAN to effectively remove occlusions, while an Attentive Residual Convolution Module (ARCM), paired with a Swin Transformer, enhances feature representation. Additionally, we introduce a Multi-Task Age Head (MTAH) that combines regression and distribution learning, further improving age estimation under occlusion. Experimental results on the FG-NET, UTKFace, and MORPH datasets demonstrate that our proposed approach surpasses existing state-of-the-art techniques for occluded facial age estimation by achieving an MAE of $3.00$, $4.54$, and $2.53$ years, respectively.

Problem

• Estimating age from occluded faces in real-world scenarios
• Removing facial occlusions using GANs and transformers
• Improving accuracy with multi-task learning for age estimation

Innovation

• GANs and transformers for occluded face age estimation
• SN-Patch GAN removes occlusions effectively
• Multi-Task Age Head combines regression and distribution learning