Skip to content

Latest commit

 

History

History
17 lines (10 loc) · 821 Bytes

README.md

File metadata and controls

17 lines (10 loc) · 821 Bytes

Scale trajectory predicted by SfM Learner

E. Chistov, M. Tregubenko, and S. Linok

Open In Colab

Overview

Visual odometry is the process of determining the position and orientation of a camera by analyzing images. One of the visual odometry problems is the discrepancy between the scale of the predicted and ground truth trajectory.

Scale discrepancy example

This repository contains algorithm to scale trajectory predicted by SfM Learner. For scaling we use metadata such as camera heigth and camera intrinsics.

Algorithm overview