A novel parametrization of the perspective-three-point problem for a direct computation of absolute camera position and orientation

By Laurent Kneip, Davide Scaramuzza, Roland Yves Siegwart. Computer Vision and Pattern Recognition, 2011

611 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/219f49c8c281e72c5c88b8a9c61ccf94a255a6ae

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

Vision-Controlled Micro Flying Robots: From System Design to Autonomous Navigation and Mapping in GPS-Denied Environments (2014, methodology): “Geometric verification is performed on the top-N most similar frames using P3P-based RANSAC [32].”
Back to the Feature: Learning Robust Camera Localization from Pixels to Pose (2021, methodology): “Such a formulation estimates the camera pose using a Perspective-n-Point (PnP) solver [1, 14, 31, 39, 40] inside a RANSAC loop [6, 19, 28, 43].”
Pose Estimation for Augmented Reality: A Hands-On Survey (2016, neutral): “, [62]) now become prevalent (especially in a SLAM context, see Section 3.”
Image-Based Localization Using LSTMs for Structured Feature Correlation (2016, methodology): “As such, camera poses are estimated using a 3-point-pose solver [25] inside a RANSAC loop [12].”
From Coarse to Fine: Robust Hierarchical Localization at Large Scale (2018, methodology): “For each place, we successively match the 2D keypoints detected in the query image to the 3D points contained in the place, and attempt to estimate a 6-DoF pose with a PnP [23] geometric consistency check within a RANSAC scheme [14].”
Benchmarking 6DOF Outdoor Visual Localization in Changing Conditions (2018, methodology): “When outliers have been discarded, three-point RANSAC [24, 32] is performed on the remaining correspondences.”
Understanding the Limitations of CNN-Based Absolute Camera Pose Regression (2019, methodology): “The pose is estimated via a P3P solver [31] inside a RANSAC [18] loop, followed by non-linear refinement of the pose [1].”
Vehicle localization using mono-camera and geo-referenced traffic signs (2015, methodology): “We apply P3P (Perspective three Point) algorithm [32] to estimate the image pose according the reconstructed 3D points.”
Revisiting the PnP Problem: A Fast, General and Optimal Solution (2013, methodology): “4, from which we can see that OPnP has the same accuracy as that of [12].”