Towards Robust Driving Perception: A Flexible Scale-Driven Family for Self-Supervised Monocular Depth Estimation

2026-07-01

AI Summary

Existing self-supervised monocular depth estimation methods suffer from limited performance in complex driving scenarios due to single-scale representations and the difficulty of modeling dynamic objects, hindering their deployment on edge devices. This work proposes FlexDepth—a scale-driven family of lightweight models that achieves efficient feature fusion and high-accuracy depth estimation through a static-dynamic decoupled two-stage training strategy and a Scale-Driven Decoder (SDD). The SDD incorporates an independent confidence assessment mechanism that enables dynamic component selection per scale, balancing accuracy and efficiency across arbitrary resolutions. Without requiring auxiliary information, FlexDepth attains state-of-the-art performance on standard driving benchmarks; its smallest variant, Flex-Nano, operates at only 0.7 GFLOPs, enabling real-time inference on mobile platforms (37.6 FPS) while maintaining strong zero-shot generalization and deployment-friendly characteristics.

Abstract

Self-Supervised Monocular Depth Estimation (MDE) has garnered attention in recent years due to its independence from ground truth. However, most existing models are limited to a single scale and exhibit considerable performance degradation in complex driving environments. Networks specifically designed to handle dynamic traffic participants tend to be overly complex, hindering their deployment on resource-constrained automotive edge devices. To address these limitations and move towards robust driving perception, we propose FlexDepth, a scale-driven and flexible family of self-supervised MDE models tailored for challenging road scenarios. FlexDepth employs a two-stage static-dynamic decoupled training strategy, enabling the independent assessment of confidence for both static backgrounds and dynamic road objects. Furthermore, it introduces a meticulously designed Scale-Driven Decoder (SDD) to dynamically select components based on scale size, facilitating efficient feature fusion and the output of high-precision depth maps. Extensive experiments on standard driving benchmarks demonstrate that without any auxiliary information, our model achieves state-of-the-art performance across arbitrary scales with minimal computational overhead. Our smallest model, Flex-Nano, requires only 0.7 GFLOPs and achieves 37.6 FPS on mobile platforms, ensuring reliable real-time perception while maintaining excellent zero-shot generalization.Our source code is avalible: https://github.com/startnew/flexdepth

Problem

• Monocular Depth Estimation
• Self-Supervised Learning
• Driving Perception
• Scale Robustness
• Edge Deployment

Innovation

• self-supervised monocular depth estimation
• scale-driven decoder
• static-dynamic decoupling
• efficient inference
• driving perception