Analytic Class Incremental Learning for Sound Source Localization with Privacy Protection

TL;DR

This paper introduces SSL-CIL, a class incremental learning method for sound source localization that maintains high accuracy without revisiting past data, ensuring privacy and adaptability in smart home applications.

Contribution

It presents a novel analytic class incremental learning approach for deep learning-based SSL that prevents catastrophic forgetting and preserves data privacy.

Findings

Achieves 90.9% localization accuracy on SSLR dataset.

Outperforms existing methods in incremental learning scenarios.

Ensures privacy by not revisiting historical data.

Abstract

Sound Source Localization (SSL) enabling technology for applications such as surveillance and robotics. While traditional Signal Processing (SP)-based SSL methods provide analytic solutions under specific signal and noise assumptions, recent Deep Learning (DL)-based methods have significantly outperformed them. However, their success depends on extensive training data and substantial computational resources. Moreover, they often rely on large-scale annotated spatial data and may struggle when adapting to evolving sound classes. To mitigate these challenges, we propose a novel Class Incremental Learning (CIL) approach, termed SSL-CIL, which avoids serious accuracy degradation due to catastrophic forgetting by incrementally updating the DL-based SSL model through a closed-form analytic solution. In particular, data privacy is ensured since the learning process does not revisit any historical data (exemplar-free), which is more suitable for smart home scenarios. Empirical results in the public SSLR dataset demonstrate the superior performance of our proposal, achieving a localization accuracy of 90.9%, surpassing other competitive methods.