Walking = Traversable? : Traversability Prediction via Multiple Human Object Tracking under Occlusion

TL;DR

This paper introduces a novel method for predicting traversability in indoor environments by using third-person monocular cameras and multiple human-object tracking, effectively handling occlusions and complex scenes.

Contribution

It replaces first-person sensors with third-person monocular cameras and combines SLAM and MOT tracking to improve traversability prediction under occlusion.

Findings

Stable traversability predictions in occluded scenarios

Effective use of SLAM and MOT for tracking

Enhanced map quality metrics for traversability maps

Abstract

The emerging ``Floor plan from human trails (PfH)" technique has great potential for improving indoor robot navigation by predicting the traversability of occluded floors. This study presents an innovative approach that replaces first-person-view sensors with a third-person-view monocular camera mounted on the observer robot. This approach can gather measurements from multiple humans, expanding its range of applications. The key idea is to use two types of trackers, SLAM and MOT, to monitor stationary objects and moving humans and assess their interactions. This method achieves stable predictions of traversability even in challenging visual scenarios, such as occlusions, nonlinear perspectives, depth uncertainty, and intersections involving multiple humans. Additionally, we extend map quality metrics to apply to traversability maps, facilitating future research. We validate our proposed method through fusion and comparison with established techniques.