Unsupervised Monocular Depth and Ego-motion Learning with Structure and Semantics

Vincent Casser, Soeren Pirk, Reza Mahjourian, Anelia Angelova

2019-06-12Motion EstimationDepth EstimationDepth And Camera MotionMonocular Depth Estimation
PaperPDF

Abstract

We present an approach which takes advantage of both structure and semantics for unsupervised monocular learning of depth and ego-motion. More specifically, we model the motion of individual objects and learn their 3D motion vector jointly with depth and ego-motion. We obtain more accurate results, especially for challenging dynamic scenes not addressed by previous approaches. This is an extended version of Casser et al. [AAAI'19]. Code and models have been open sourced at https://sites.google.com/corp/view/struct2depth.

Results

TaskDatasetMetricValueModel
Depth EstimationKITTI Eigen split unsupervisedabsolute relative error0.1412Struct2Depth M
3DKITTI Eigen split unsupervisedabsolute relative error0.1412Struct2Depth M

Related Papers

DINO-VO: A Feature-based Visual Odometry Leveraging a Visual Foundation Model2025-07-17$S^2M^2$: Scalable Stereo Matching Model for Reliable Depth Estimation2025-07-17$π^3$: Scalable Permutation-Equivariant Visual Geometry Learning2025-07-17Efficient Calisthenics Skills Classification through Foreground Instance Selection and Depth Estimation2025-07-16Vision-based Perception for Autonomous Vehicles in Obstacle Avoidance Scenarios2025-07-16MonoMVSNet: Monocular Priors Guided Multi-View Stereo Network2025-07-15Towards Depth Foundation Model: Recent Trends in Vision-Based Depth Estimation2025-07-15Cameras as Relative Positional Encoding2025-07-14