# TROVE Feature Detection for Online Pose Recovery by Binocular Cameras

## TL;DR

This paper introduces a real-time method for estimating 6-DoF ego-states using TROVE features, which are common in man-made environments, enabling accurate indoor localization without traditional corner detection or PnP methods.

## Contribution

The paper presents a novel TROVE-based feature detection approach that enables fast and accurate pose estimation without relying on conventional corner features or PnP algorithms.

## Key findings

- Achieves real-time pose estimation up to 60 Hz.
- Attitude accuracy reaches 0.3 degrees.
- Position accuracy reaches 2 cm indoors.

## Abstract

This paper proposes a new and efficient method to estimate 6-DoF ego-states: attitudes and positions in real time. The proposed method extract information of ego-states by observing a feature called "TROVE" (Three Rays and One VErtex). TROVE features are projected from structures that are ubiquitous on man-made constructions and objects. The proposed method does not search for conventional corner-type features nor use Perspective-n-Point (PnP) methods, and it achieves a real-time estimation of attitudes and positions up to 60 Hz. The accuracy of attitude estimates can reach 0.3 degrees and that of position estimates can reach 2 cm in an indoor environment. The result shows a promising approach for unmanned robots to localize in an environment that is rich in man-made structures.