1,576 Hits in 5.8 sec

Scene Analysis for Service Robots [chapter]

Robert Eidenberger, Thilo Grundmann, Martin Schneider, Wendelin Feiten, Michael Fiegert, Georg v. Wichert, Gisbert Lawitzky
2012 Springer Tracts in Advanced Robotics  
A scene analysis module for service robots is presented which uses SIFT in a stereo setting, a systematic handling of uncertainties and an active perception component.  ...  The system is integrated and evaluated on the DESIRE two-arm mobile robot. Complex everyday scenes composed of various items from a 100-object database are analyzed successfully and efficiently.  ...  One of the earlier papers on object pose estimation [12] dealt already with the high number of one hundred different classes.  ... 
doi:10.1007/978-3-642-25116-0_15 fatcat:5x3ls6ethrd6rjrf46dbh4ohci

A Comprehensive Review on 3D Object Detection and 6D Pose Estimation with Deep Learning

Sabera Hoque, MD. Yasir Arafat, Shuxiang Xu, Ananda Maiti, Yuchen Wei
2021 IEEE Access  
Chabot [27] made a framework called Deep MANTA for 3D object detection based on a 825 single-dimensional image in an end-to-end fashion network, determining the object class, 2D region proposal generation  ...  It is easier to predict the 6D pose in RGBD images than in 325 RGB images because the 6D pose is a complex combination of 3D rotation of an object (raw, pitch, yaw) and 3D coordinates (X, Y, Z) at the  ... 
doi:10.1109/access.2021.3114399 fatcat:kvdwsslqxff3lkh27tsdsciqma

Pixels, Stixels, and Objects [chapter]

David Pfeiffer, Friedrich Erbs, Uwe Franke
2012 Lecture Notes in Computer Science  
The high spatial and temporal resolution allows for robust obstacle detection in complex inner city scenarios, including pedestrian recognition and detection of partially hidden moving objects.  ...  This paper shows how this representation forms the foundation for a very efficient, robust and comprehensive understanding of traffic scenes.  ...  Introduction Driver assistance in complex urban environments is a challenging task requiring comprehensive perception of static and moving objects.  ... 
doi:10.1007/978-3-642-33885-4_1 fatcat:fgchjdk4frfxthw36hwvire4su

W-PoseNet: Dense Correspondence Regularized Pixel Pair Pose Regression [article]

Zelin Xu, Ke Chen, Kui Jia
2021 arXiv   pre-print
In other words, local features learned for pose regression in our deep network are regularized by explicitly learning pixel-wise correspondence mapping onto 3D pose-sensitive coordinates as an auxiliary  ...  This paper introduces a novel pose estimation algorithm W-PoseNet, which densely regresses from input data to 6D pose and also 3D coordinates in model space.  ...  In light of this, encoding a discriminative and robust feature representation for each instance is essential for predicting its 6D pose.  ... 
arXiv:1912.11888v2 fatcat:j72va3fqpbafxfh642gciqdu6u

Image-based camera localization: an overview

Yihong Wu, Fulin Tang, Heping Li
2018 Visual Computing for Industry, Biomedicine, and Art  
Trends for future development are also discussed. This will be useful not only to researchers, but also to engineers and other individuals interested in this field.  ...  However, there has not been a complete review on image-based camera localization. It is urgent to map this topic to enable individuals enter the field quickly.  ...  Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Received: 3 April 2018 Accepted: 6 July 2018  ... 
doi:10.1186/s42492-018-0008-z pmid:32240389 fatcat:3ojnmancw5fbjcdhasko2mz4cq

Planning for Multi-stage Forceful Manipulation [article]

Rachel Holladay, Tomás Lozano-Pérez, Alberto Rodriguez
2021 arXiv   pre-print
In two domains, opening a childproof bottle and twisting a nut, we demonstrate how the system considers a combinatorial number of strategies and how choosing actions that are robust to parameter variations  ...  Multi-stage forceful manipulation tasks, such as twisting a nut on a bolt, require reasoning over interlocking constraints over discrete as well as continuous choices.  ...  We thank the members of the MCube Lab and LIS Group for their helpful feedback and especially Caelan Garrett for the many PDDLStream discussions.  ... 
arXiv:2101.02679v2 fatcat:5my5zgqozfe6dbnjay7v5xmewa

Mixing deep learning with classical vision for object recognition

Maciej Stefańczyk, Tomasz Bocheński
2020 Journal of WSCG  
In this paper, we try to show the advantages of mixing deep learning object recognition/detection with classical 6-DOF pose estimation algorithms, with a focus on applications in service robotics.  ...  Accurate 6-DOF object localization in the image is a more laborious task and requires more complex training data to be available.  ...  PoseCNN, SSD-6D, and similar requires pose-labeled images as the input for the training.  ... 
doi:10.24132/jwscg.2020.28.18 fatcat:2gox4lcdqzc6fct2dpk4hnj7rq

Multi-path Learning for Object Pose Estimation Across Domains [article]

Martin Sundermeyer, Maximilian Durner, En Yen Puang, Zoltan-Csaba Marton, Narunas Vaskevicius, Kai O. Arras, Rudolph Triebel
2020 arXiv   pre-print
We introduce a scalable approach for object pose estimation trained on simulated RGB views of multiple 3D models together.  ...  Despite training jointly on multiple objects, our 6D Object Detection pipeline achieves state-of-the-art results on T-LESS at much lower runtimes than competing approaches.  ...  search in the large 6D pose space.  ... 
arXiv:1908.00151v2 fatcat:nfspzncnxveilcjncfhswnqvgu

SPEED+: Next-Generation Dataset for Spacecraft Pose Estimation across Domain Gap [article]

Tae Ha Park, Marcus Märtens, Gurvan Lecuyer, Dario Izzo, Simone D'Amico
2021 arXiv   pre-print
dataset of images of the intended target in the space environment.  ...  SPEED+ is used in the second international Satellite Pose Estimation Challenge co-hosted by SLAB and the Advanced Concepts Team of the European Space Agency to evaluate and compare the robustness of spaceborne  ...  Finally, HigherHRNet is trained to detect the same 11 keypoints as in KRN, but it returns heatmaps associated with each keypoints. Then, PnP is solved for complete 6D pose.  ... 
arXiv:2110.03101v2 fatcat:q2gzjwhcdvdjvcbdu2y6sf7ovu

Learning visual representations for perception-action systems

Justus Piater, Sébastien Jodogne, Renaud Detry, Dirk Kraft, Norbert Krüger, Oliver Kroemer, Jan Peters
2010 The international journal of robotics research  
In contrast to the minimalistic visual representations produced by RLVC and RLJC, our second method learns structural object models for robust object detection and pose estimation by probabilistic inference  ...  We discuss vision as a sensory modality for systems that effect actions in response to perceptions.  ...  We thank Volker Krüger and Dennis Herzog for their support during the recording of the human demonstration data.  ... 
doi:10.1177/0278364910382464 fatcat:rxvavyfjfjfs5ht7g23w5sawoe

SLAM with Objects using a Nonparametric Pose Graph [article]

Beipeng Mu, Shih-Yuan Liu, Liam Paull, John Leonard, Jonathan How
2017 arXiv   pre-print
Mapping and self-localization in unknown environments are fundamental capabilities in many robotic applications.  ...  This paper proposes a novel nonparametric pose graph that models data association and SLAM in a single framework.  ...  ACKNOWLEDGMENTS This research is supported in part by ARO MURI grant W911NF-11-1-0391, ONR grant N00014-11-1-0688 and NSF Award IIS-1318392.  ... 
arXiv:1704.05959v1 fatcat:pnjf53sddvhiljdlrlojepo5oq

Probabilistic Algorithms in Robotics

Sebastian Thrun
2000 The AI Magazine  
The views and conclusions contained in this document are those of the author and should not be interpreted as necessarily representing official policies or endorsements, either expressed or implied, of  ...  Another class of approaches relies on tree search for policy determination, such as the work on active perception and sensor planning by Kristensen (1997 Kristensen ( , 1996 .  ...  In all these cases, the probabilistic nature of the algorithms has been essential for achieving robustness in the face of uncertain and dynam-ic environments.  ... 
doi:10.1609/aimag.v21i4.1534 dblp:journals/aim/Thrun00 fatcat:jm6mxmuc4rh6ngtdrcw3czsyca

Learning Visual Representations for Interactive Systems [chapter]

Justus Piater, Sébastien Jodogne, Renaud Detry, Dirk Kraft, Norbert Krüger, Oliver Krömer, Jan Peters
2011 Springer Tracts in Advanced Robotics  
In contrast to the minimalistic visual representations produced by RLVC and RLJC, our second method learns structural object models for robust object detection and pose estimation by probabilistic inference  ...  We describe two quite different methods for associating action parameters to visual percepts. Our RLVC algorithm performs reinforcement learning directly on the visual input space.  ...  We thank Volker Krüger and Dennis Herzog for their support during the recording of the human demonstration data.  ... 
doi:10.1007/978-3-642-19457-3_24 fatcat:gioekjwsufap3mkoahflly2zqi

Comparing ICP variants on real-world data sets

François Pomerleau, Francis Colas, Roland Siegwart, Stéphane Magnenat
2013 Autonomous Robots  
Because ICP has many variants whose performances depend on the environment and the sensor, hundreds of variations have been published.  ...  Besides demonstrating the need for improved ICP methods for natural, unstructured and information-deprived environments, these baseline variants also provide a solid basis to which novel solutions could  ...  Acknowledgments This work was supported by the EU FP7 IP projects Natural Human-Robot Cooperation in Dynamic Environments (ICT-247870) and myCopter (FP7-AAT-2010-RTD-1). F.  ... 
doi:10.1007/s10514-013-9327-2 fatcat:lu6n2drpr5fqfb653ybkmtvefu

Closing the Loop for Robotic Grasping: A Real-time, Generative Grasp Synthesis Approach [article]

Douglas Morrison, Peter Corke, Jürgen Leitner
2018 arXiv   pre-print
The light-weight and single-pass generative nature of our GG-CNN allows for closed-loop control at up to 50Hz, enabling accurate grasping in non-static environments where objects move and in the presence  ...  In our real-world tests, we achieve an 83% grasp success rate on a set of previously unseen objects with adversarial geometry and 88% on a set of household objects that are moved during the grasp attempt  ...  ACKNOWLEDGMENTS This research was supported by the Australian Research Council Centre of Excellence for Robotic Vision (project number CE140100016).  ... 
arXiv:1804.05172v2 fatcat:xwzi57et7fg3rofqey5esivuda
« Previous Showing results 1 — 15 out of 1,576 results