SVO: Fast semi-direct monocular visual odometry

TLDR

The authors propose a semi‑direct monocular visual odometry algorithm that is precise, robust, and faster than current state‑of‑the‑art methods. The algorithm uses a semi‑direct approach that operates directly on pixel intensities, eliminates costly feature extraction, employs probabilistic mapping to model outliers, and achieves high‑speed state estimation (55 fps on embedded hardware, >300 fps on a laptop). The method delivers precise, high‑frame‑rate motion estimation that improves robustness in low‑texture, repetitive, and high‑frequency scenes, and the authors release the SVO implementation as open‑source software.

Abstract

We propose a semi-direct monocular visual odometry algorithm that is precise, robust, and faster than current state-of-the-art methods. The semi-direct approach eliminates the need of costly feature extraction and robust matching techniques for motion estimation. Our algorithm operates directly on pixel intensities, which results in subpixel precision at high frame-rates. A probabilistic mapping method that explicitly models outlier measurements is used to estimate 3D points, which results in fewer outliers and more reliable points. Precise and high frame-rate motion estimation brings increased robustness in scenes of little, repetitive, and high-frequency texture. The algorithm is applied to micro-aerial-vehicle state-estimation in GPS-denied environments and runs at 55 frames per second on the onboard embedded computer and at more than 300 frames per second on a consumer laptop. We call our approach SVO (Semi-direct Visual Odometry) and release our implementation as open-source software.