Elderly Activity Recognition in the Wild: Results from the EAR Challenge

2025-03-10

AI Summary

This work addresses the real-world task of recognizing Activities of Daily Living (ADL) for elderly individuals, focusing on fine-grained classification of six distinct actions. To tackle cross-domain distribution shift and poor generalization under limited labeled data—common challenges in geriatric activity recognition—we propose a collaborative dataset construction and bias-correction strategy specifically tailored to older adults. Our method adopts Temporal Shift Module (TSM) as the backbone architecture, integrates multiple public video datasets, and incorporates domain-informed video preprocessing and transfer-learning-based fine-tuning. Evaluated on the EAR Challenge public leaderboard, the proposed approach achieves an accuracy of 0.81455, substantially outperforming baseline models. The implementation is fully open-sourced, providing a reproducible, robust, and deployable solution for elderly behavior understanding in practical settings.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Computer Vision: Video Understanding & Activity Analysis
• Application Domains: Humanities & Computational Social Science

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

This paper presents our solution for the Elderly Action Recognition (EAR) Challenge, part of the Computer Vision for Smalls Workshop at WACV 2025. The competition focuses on recognizing Activities of Daily Living (ADLs) performed by the elderly, covering six action categories with a diverse dataset. Our approach builds upon a state-of-the-art action recognition model, fine-tuned through transfer learning on elderly-specific datasets to enhance adaptability. To improve generalization and mitigate dataset bias, we carefully curated training data from multiple publicly available sources and applied targeted pre-processing techniques. Our solution currently achieves 0.81455 accuracy on the public leaderboard, highlighting its effectiveness in classifying elderly activities. Source codes are publicly available at https://github.com/ffyyytt/EAR-WACV25-DAKiet-TSM.

Problem

• Recognizing elderly daily activities in diverse environments
• Enhancing action recognition models for elderly-specific datasets
• Improving generalization and reducing bias in activity classification

Innovation

• Fine-tuned state-of-the-art action recognition model
• Curated diverse training data for generalization
• Applied targeted pre-processing to reduce bias