QuadricSLAM: Dual Quadrics From Object Detections as Landmarks in Object-Oriented SLAM

By Lachlan Nicholson, Michael Milford, Niko Sünderhauf. IEEE Robotics and Automation Letters, 2018

341 citations recorded by the source corpus.

In this letter, we use two-dimensional (2-D) object detections from multiple views to simultaneously estimate a 3-D quadric surface for each object and localize the camera position. We derive a simultaneous localization and mapping (SLAM) formulation that uses dual quadrics as 3-D landmark representations, exploiting their ability to compactly represent the size, position and orientation of an object, and show how 2-D object detections can directly constrain the quadric parameters via a novel geometric error formulation. We develop a sensor model for object detectors that addresses the challenge of partially visible objects, and demonstrate how to jointly estimate the camera pose and constrained dual quadric parameters in factor graph based SLAM with a general perspective camera.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/a104793d8e002a254dc861f78aa238d9096cc59f

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Aerospace Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

Aligning Cyber Space With Physical World: A Comprehensive Survey on Embodied AI (2024, neutral): “Embodiedperceptionrequires visualperceptionandreasoning,understandingthe3-Drelations within a scene, and predicting and performing complex tasks based on visual information.”
A Review and Comparative Study on Probabilistic Object Detection in Autonomous Driving (2020, positive): “Another task that could beneﬁt from uncertainty estimates from object detection is the object-based simultaneous localization and mapping [117], where the predicted mean and covariance matrices from bounding boxes can be used as measurement updates.”
KITTI-360: A Novel Dataset and Benchmarks for Urban Scene Understanding in 2D and 3D (2021, background): “One possibility for improving localization accuracy is to exploit the 3D bounding boxes and instance labels available in our dataset [71], [109].”
A Survey on Active Simultaneous Localization and Mapping: State of the Art and New Frontiers (2022, background): “Instead of geometric features, a sparse map can capture objects, described by a semantic category, pose, and shape [91], [92].”
CubeSLAM: Monocular 3-D Object SLAM (2018, methodology): “Rubino et al. [24] solved multi-view 3D ellipsoid object localization analytically and QuadriSLAM [25] extended it to an online SLAM without prior models.”
Kimera: From SLAM to spatial perception with 3D dynamic scene graphs (2021, methodology): “Object-based approaches compute an object map and include SLAM++ (Salas-Moreno et al. 2013), XIVO (Dong et al. 2017), OrcVIO (Shan et al. 2019), QuadricSLAM (Nicholson et al. 2018), and (Bowman et al. 2017).”
3D Dynamic Scene Graphs: Actionable Spatial Perception with Places, Objects, and Humans (2020, background): “Object-based approaches compute an object map and include SLAM++ [93], XIVO [21], OrcVIO [95], QuadricSLAM [72], and [12].”
Hydra: A Real-time Spatial Perception System for 3D Scene Graph Construction and Optimization (2022, neutral): “The literature has focused on both object-based maps [10, 14, 41, 42, 53, 56] and dense maps, including volumetric models [23, 38, 40], point clouds [6, 32, 61], and 3D meshes [48, 51].”