Fractal Markers: A New Approach for Long-Range Marker Pose Estimation Under Occlusion

Abstract

Squared fiducial markers are a powerful tool for camera pose estimation in applications such as robots, unmanned vehicles and augmented reality. The four corners of a single marker are enough to estimate the pose of a calibrated camera. However, they have some limitations. First, the methods proposed for detection are ineffective under occlusion. A small occlusion in any part of the marker makes it undetectable. Second, the range at which they can be detected is limited by their size. Very big markers can be detected from a far distance, but as the camera approaches them, they are not fully visible, and thus they can not be detected. Small markers, however, can not be detected from large distances. This paper proposes solutions to the above-mentioned problems. We propose the Fractal Marker, a novel type of marker that is built as an aggregation of squared markers, one into another, in a recursive manner. Also, we proposed a novel method for detecting Fractal Markers under severe occlusions. The results of our experiments show that the proposed method achieves a wider detection range than traditional markers and great robustness to occlusion.

Designing a Simple Fiducial Marker for Localization in Spatial Scenes Using Neural Networks

“…The authors focus on a hierarchical design that would allow the nesting of multiple markers. Similar solutions were proposed by Tateno et al [ 69 ] in 2007 and in a system called Fractal Markers [ 70 ] in 2019. However, these systems are always nesting one marker inside another single marker.…”

Error Reduction in Vision-Based Multirotor Landing System

“…The error position for each of the components is given by Equation (21). Looking at the errors (Figure 9), the error distribution increases with increasing distance from the north-east plane origin (0, 0).…”

“…Regarding pattern recognition landing systems, works such as [13,14], among others [15][16][17][18], focus on finding the position using known patterns by Perspective-n-Point (PnP) algorithms [19]. Patterns such as Aruco [20], charuco, or new fractal patterns such as [21] or the deep learning trend You Only Look Once (YOLO) [22] try to improve the pattern pose estimation and prove to be widespread systems in the literature. In the literature and throughout this paper, an object "pose" means a set of position and heading of a specific reference system.…”

An Onboard Vision-Based System for Autonomous Landing of a Low-Cost Quadrotor on a Novel Landing Pad

“…In [36], the proposed Fractal Marker, a novel type of marker, is built as an aggregation of squared markers of different sizes, one into another, in a recursive manner, and a method for marker detection under severe occlusions is also presented. Taking into account the latest advances in the fractal ArUco, a novel landing pad with a wider detection range is designed in this paper to achieve a better performance in autonomous landing tasks.…”