Filters








269,430 Hits in 2.8 sec

Robust Optimal Pose Estimation [chapter]

Olof Enqvist, Fredrik Kahl
2008 Lecture Notes in Computer Science  
A common problem in camera pose estimation is the existence of false correspondences between image features and modeled 3D points.  ...  We study the problem of estimating the position and orientation of a calibrated camera from an image of a known scene.  ...  This is the key component for the main contribution of the paper: an efficient, robust and optimal solution to the pose problem.  ... 
doi:10.1007/978-3-540-88682-2_12 fatcat:ynvztbgjozewtmnkfwb2bux2ry

RNNPose: Recurrent 6-DoF Object Pose Refinement with Robust Correspondence Field Estimation and Pose Optimization [article]

Yan Xu, Kwan-Yee Lin, Guofeng Zhang, Xiaogang Wang, Hongsheng Li
2022 arXiv   pre-print
correspondences during pose optimization.  ...  In this paper, we propose a framework based on a recurrent neural network (RNN) for object pose refinement, which is robust to erroneous initial poses and occlusions.  ...  The correspondence field estimation and pose optimization rely on each other and improve recurrently for robust pose refinement.  ... 
arXiv:2203.12870v3 fatcat:62dgeycm6zdrnduvojvb7usvsm

Hybrid Inference Optimization for robust pose graph estimation

Aleksandr V. Segal, Ian D. Reid
2014 2014 IEEE/RSJ International Conference on Intelligent Robots and Systems  
In this paper we introduce a new optimization algorithm for networks of switched nonlinear objectives and apply this to the important problem of pose graph estimation for robot localization and mapping  ...  We apply the new algorithm to the problem of robust pose graph estimation in the presence of incorrect loop closures and compare against three recently published approaches to the same problem.  ...  This has motivated research on robust pose graph optimization algorithms which go beyond classical robust estimation techniques and deal with the loop closures directly in the back end.  ... 
doi:10.1109/iros.2014.6942928 dblp:conf/iros/SegalR14 fatcat:b6c7tgxvq5beji3dsgvrdmdpwu

Robust binocular pose estimation based on pigeon-inspired optimization

Lu Gan, Haibin Duan
2015 2015 IEEE 10th Conference on Industrial Electronics and Applications (ICIEA)  
In this paper, an accurate and robust pose estimation algorithm for binocular camera systems is proposed based on pigeon-inspired optimization (PIO), which can be easily generalized to estimation for multiple  ...  Keywords-pose estimation; binocular camera systems; multiple camera systems (MCS); pigeon-inspired optimization (PIO)  ...  CONCLUSION This paper presented a robust and accurate pose estimation algorithm for binocular camera systems, which can be also used to estimate the pose of MCS.  ... 
doi:10.1109/iciea.2015.7334261 fatcat:twjgqvfai5cuxgkrtgr7fqhs4q

Certifiably Optimal and Robust Camera Pose Estimation from Points and Lines

Lei Sun, Zhongliang Deng
2020 IEEE Access  
The Perspective-n-Point-and-Line (PnPL) problem, as a cornerstone in geometric computer vision, seeks to estimate the absolute pose of a calibrated camera from 3D-to-2D point and line correspondences.  ...  In this paper, we present a certifiably globally optimal and robust solution to the PnPL problem with a large number of outliers among the correspondences.  ...  Pose estimation based on real data [46] with 40% outliers.  ... 
doi:10.1109/access.2020.3006056 fatcat:tq2tzx432zhhvio4dl5sgolgvy

Optimal and Robust Category-level Perception: Object Pose and Shape Estimation from 2D and 3D Semantic Keypoints [article]

Jingnan Shi, Heng Yang, Luca Carlone
2022 arXiv   pre-print
and shape are estimated from 2D or 3D keypoints via non-convex optimization.  ...  Our first contribution is to develop PACE3D* and PACE2D*, the first certifiably optimal solvers for pose and shape estimation using 3D and 2D keypoints, respectively.  ...  We address the shortcomings of existing approaches for pose and shape estimation based on the active shape model and propose the first approaches that can compute optimal estimates and are robust to outliers  ... 
arXiv:2206.12498v1 fatcat:3l5s6tfvxfdr7ggyvilm4gbodi

Pose Estimation for Robot Manipulators via Keypoint Optimization and Sim-to-Real Transfer [article]

Jingpei Lu, Florian Richter, Michael Yip
2022 arXiv   pre-print
Keypoint detection is an essential building block for many robotic applications like motion capture and pose estimation.  ...  The approach involves finding the optimal set of keypoints on robotic manipulators for robust visual detection and localization.  ...  The Algorithm 1 is applied to optimize the set of 7 (K = 7) keypoints for robust detection.  ... 
arXiv:2010.08054v3 fatcat:usw3t2ajtrbqdnmqozt32sgtv4

An Assessment of Nonlinear Optimization and Speeded up Robust Features (SURF) Algorithm for Estimating Object Space Transformation Parameters for UAV Pose Estimation

Bassem Sheta, Mohamed Elhabiby, Naser El-Sheimy
2012 GEOMATICA  
Five different nonlinear optimization methods are presented for estimating the transformation parameters -these include four gradient based nonlinear optimization methods; Trust region, Trust region dogleg  ...  Although a robust scale and rotation invariant image matching algorithm is important for VBN of aerial vehicles, the proper estimation of the collinearity equation object space transformation parameters  ...  Dans ce cas, les approches traditionnelles des moindres carrés échoueront ou prendront plus de temps à estimer les paramètres de transformation entre l'objet et sa position spatiale en raison de la non-linéarité  ... 
doi:10.5623/cig2012-056 fatcat:xyuwpwztgzgx7i7xuszo2c64ee

SLAM-Supported Self-Training for 6D Object Pose Estimation [article]

Ziqi Lu, Yihao Zhang, Kevin Doherty, Odin Severinsen, Ethan Yang, John Leonard
2022 arXiv   pre-print
An automatic covariance tuning method is developed for robust PGO by allowing the measurement uncertainty models to change as part of the optimization process.  ...  Combining the network predictions with robot odometry, we can build a consistent object-level environment map via pose graph optimization (PGO).  ...  III.C) to obtain robust SLAM estimates {x t }, { ˆ j }. 2 Our method is not restricted to using a particular pose estimator. B.  ... 
arXiv:2203.04424v2 fatcat:kwuvnb7jr5ai5ib2fahqoizbe4

Robust direct vision-based pose tracking using normalized mutual information (Addendum)

Hang Luo, Christian Pape, Eduard Reithmeier, Sen Han, Toru Yoshizawa, Song Zhang
2019 Optical Metrology and Inspection for Industrial Applications V  
Both publications are concerned with robust pose estimation methods for the task of direct visual odometry.  ...  We use a re-weighting SSD-based tracker to get a coarse pose estimate on lower resolution levels.  ...  Afterwards, a NMI maximization on high-resolution levels is performed to optimize the coarse estimate. The implementation of the information-based optimizer contains another difference.  ... 
doi:10.1117/12.2534586 fatcat:q4rle7pn5jgqpmst34iro2ktmi

Accurate, robust, and real-time estimation of finger pose with a motion capture system

Youngmok Yun, Priyanshu Agarwal, Ashish D. Deshpande
2013 2013 IEEE/RSJ International Conference on Intelligent Robots and Systems  
Finger exoskeletons, haptic devices, and augmented reality applications demand an accurate, robust, and fast estimation of finger pose.  ...  We present a novel finger pose estimation method using a motion capture system. The method combines system identification and state estimation in a unified framework.  ...  This two-stage method realizes accurate, robust, and highfrequency estimation of finger pose with a motion capture system.  ... 
doi:10.1109/iros.2013.6696567 dblp:conf/iros/YunAD13 fatcat:av5nuweskzcj7egeiksyqn6de4

Kimera-Multi: Robust, Distributed, Dense Metric-Semantic SLAM for Multi-Robot Systems [article]

Yulun Tian, Yun Chang, Fernando Herrera Arias, Carlos Nieto-Granda, Jonathan P. How, Luca Carlone
2021 arXiv   pre-print
When communication is available, robots initiate a distributed place recognition and robust pose graph optimization protocol based on a novel distributed graduated non-convexity algorithm.  ...  The proposed protocol allows the robots to improve their local trajectory estimates by leveraging inter-robot loop closures while being robust to outliers.  ...  Robust Distributed Pose-Graph Optimization Following the initialization stage, robots perform robust distributed PGO to obtain optimal trajectory estimates while simultaneously rejecting outlier loop closures  ... 
arXiv:2106.14386v2 fatcat:2k3x4c6nqzdnvfxix4zkf67xce

A Novel Method for the Absolute Pose Problem with Pairwise Constraints [article]

Yinlong Liu, Xuechen Li, Manning Wang, Guang Chen, Zhijian Song, Alois Knoll
2019 arXiv   pre-print
In this paper, we consider pairwise constraints and propose a globally optimal algorithm for solving the absolute pose estimation problem.  ...  However, absolute pose estimation is a geometric parameter estimation problem, and thus has special constraints.  ...  In this body of work, the pose estimation problem is formulated as a robust optimization problem.  ... 
arXiv:1903.10175v2 fatcat:2ijwreb5rjh7pa2aqjjnnpp6mi

Integation Methods of Model-Free Features for 3D Tracking [chapter]

Ville Kyrki, Kerstin Schmock
2005 Lecture Notes in Computer Science  
The relative pose change between frames is estimated using an optimization approach. This allows the pose change to be integrated very efficiently in a Kalman filter.  ...  Our first approach estimates the pose change in a least squares sense while the second one uses M-estimators to decrease the effect of outliers.  ...  Robustness The robustness of the two optimization based pose tracking methods was inspected by ignoring totally the model-based cue.  ... 
doi:10.1007/11499145_57 fatcat:leeryj5mfjcyfic4df2lowlfne

NeMo: Neural Mesh Models of Contrastive Features for Robust 3D Pose Estimation [article]

Angtian Wang, Adam Kortylewski, Alan Yuille
2021 arXiv   pre-print
In this work, we show that standard deep learning approaches to 3D pose estimation are not robust when objects are partially occluded or viewed from a previously unseen pose.  ...  3D pose estimation is a challenging but important task in computer vision.  ...  ROBUST 3D POSE ESTIMATION UNDER OCCLUSION Baseline performances.  ... 
arXiv:2101.12378v3 fatcat:g2dknojgp5agpkckshpxpg3lmy
« Previous Showing results 1 — 15 out of 269,430 results