Lifelong-MonoDepth: Lifelong Learning for Multi-Domain Monocular Metric Depth Estimation

TL;DR

Lifelong-MonoDepth introduces a multi-head framework and uncertainty-aware learning to improve monocular depth estimation across multiple domains, addressing scale variation and domain gaps for autonomous systems.

Contribution

The paper proposes a novel lightweight multi-head model, an uncertainty-aware lifelong learning strategy, and an online domain predictor for robust multi-domain depth estimation.

Findings

Achieves 8% to 15% improvement over benchmarks.

Effectively handles significant domain gaps and scale variations.

Demonstrates stability, efficiency, and plasticity in experiments.

Abstract

With the rapid advancements in autonomous driving and robot navigation, there is a growing demand for lifelong learning models capable of estimating metric (absolute) depth. Lifelong learning approaches potentially offer significant cost savings in terms of model training, data storage, and collection. However, the quality of RGB images and depth maps is sensor-dependent, and depth maps in the real world exhibit domain-specific characteristics, leading to variations in depth ranges. These challenges limit existing methods to lifelong learning scenarios with small domain gaps and relative depth map estimation. To facilitate lifelong metric depth learning, we identify three crucial technical challenges that require attention: i) developing a model capable of addressing the depth scale variation through scale-aware depth learning, ii) devising an effective learning strategy to handle significant domain gaps, and iii) creating an automated solution for domain-aware depth inference in practical applications. Based on the aforementioned considerations, in this paper, we present i) a lightweight multi-head framework that effectively tackles the depth scale imbalance, ii) an uncertainty-aware lifelong learning solution that adeptly handles significant domain gaps, and iii) an online domain-specific predictor selection method for real-time inference. Through extensive numerical studies, we show that the proposed method can achieve good efficiency, stability, and plasticity, leading the benchmarks by 8% to 15%.