Lifelong Change Detection: Continuous Domain Adaptation for Small Object Change Detection in Every Robot Navigation

TL;DR

This paper introduces a self-supervised lifelong learning method for small object change detection in robot navigation, leveraging detected changes as priors to improve future detection without manual annotations.

Contribution

It proposes a novel self-supervised framework that continuously adapts to new environments for ground-view small object change detection in robotics.

Findings

Effective in practical robot navigation scenarios

Improves change detection accuracy over time

Operates without manual annotations

Abstract

The recently emerging research area in robotics, ground view change detection, suffers from its ill-posed-ness because of visual uncertainty combined with complex nonlinear perspective projection. To regularize the ill-posed-ness, the commonly applied supervised learning methods (e.g., CSCD-Net) rely on manually annotated high-quality object-class-specific priors. In this work, we consider general application domains where no manual annotation is available and present a fully self-supervised approach. The present approach adopts the powerful and versatile idea that object changes detected during everyday robot navigation can be reused as additional priors to improve future change detection tasks. Furthermore, a robustified framework is implemented and verified experimentally in a new challenging practical application scenario: ground-view small object change detection.