Self-Supervised Monocular Depth Estimation: Solving the Dynamic Object Problem by Semantic Guidance

TL;DR

This paper introduces SGDepth, a self-supervised monocular depth estimation method that effectively handles moving dynamic objects by integrating semantic segmentation, semantic masking, and detection techniques, improving accuracy without requiring depth labels.

Contribution

The work presents a novel semantically-guided approach that jointly trains depth and segmentation, and employs semantic masking and detection to address dynamic objects in self-supervised depth estimation.

Findings

Outperforms all baselines on the Eigen split benchmark.

Effectively handles moving dynamic objects in depth estimation.

No test-time refinement needed for superior performance.

Abstract

Self-supervised monocular depth estimation presents a powerful method to obtain 3D scene information from single camera images, which is trainable on arbitrary image sequences without requiring depth labels, e.g., from a LiDAR sensor. In this work we present a new self-supervised semantically-guided depth estimation (SGDepth) method to deal with moving dynamic-class (DC) objects, such as moving cars and pedestrians, which violate the static-world assumptions typically made during training of such models. Specifically, we propose (i) mutually beneficial cross-domain training of (supervised) semantic segmentation and self-supervised depth estimation with task-specific network heads, (ii) a semantic masking scheme providing guidance to prevent moving DC objects from contaminating the photometric loss, and (iii) a detection method for frames with non-moving DC objects, from which the depth of DC objects can be learned. We demonstrate the performance of our method on several benchmarks, in particular on the Eigen split, where we exceed all baselines without test-time refinement.