Filters








3,536 Hits in 4.2 sec

Probabilistic Active Recognition of Multiple Objects Using Hough-Based Geometric Matching Features [chapter]

Natasha Govender, Philip Torr, Mogomotsi Keaikitse, Fred Nicolls, Jonathan Warrell
2015 Cognitive Systems Monographs  
Further, we investigate an efficient active viewpoint selection algorithm based on vocabulary-tree clustering and Term Frequency Inverse Document Frequency (TFIDF) uniqueness metric.  ...  Further, we show our viewpoint selection algorithm to be both faster and more accurate than alternatives in both Bayesian and discriminative contexts, including methods based on mutual information in the  ...  Further, we investigate an efficient active viewpoint selection algorithm based on vocabulary-tree clustering and Term Frequency Inverse Document Frequency (TFIDF) uniqueness metric.  ... 
doi:10.1007/978-3-662-43859-6_6 fatcat:mfyk5nrkszfudklldbmoxqcsze

Active object recognition by view integration and reinforcement learning

Lucas Paletta, Axel Pinz
2000 Robotics and Autonomous Systems  
Active recognition of three-dimensional objects involves the observer in a search for discriminative evidence, e.g., by change of its viewpoint.  ...  The information gain in fusing successive object hypotheses provides a utility measure to reinforce actions leading to discriminative viewpoints.  ...  We wish to thank Manfred Prantl and Hermann Borotschnig for useful discussions.  ... 
doi:10.1016/s0921-8890(99)00079-2 fatcat:mmdabedtinettm3qi34vf7dawe

Active object recognition using vocabulary trees

N. Govender, J. Claassens, F. Nicolls, J. Warrell
2013 2013 IEEE Workshop on Robot Vision (WORV)  
This paper presents an efficient feature-based active vision system for the recognition and verification of objects that are occluded, appear in cluttered scenes and may be visually similar to other objects  ...  This system is designed using a selector-observer framework where the selector is responsible for the automatic selection of the next best viewpoint and a Bayesian 'observer' updates the belief hypothesis  ...  Active Viewpoint Selection The aim of the automatic view selection algorithm is to select the 'next best viewpoint' for object recognition and verification i.e. the viewpoint which will provide the most  ... 
doi:10.1109/worv.2013.6521945 fatcat:uhremxipm5elbnoume5hhqdt5i

Fast parametric viewpoint estimation for active object detection

Robert Eidenberger, Thilo Grundmann, Wendelin Feiten, Raoul Zoellner
2008 2008 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems  
Most current solutions to active perception planning struggle with complex state representations or fast and efficient sensor parameter selection strategies.  ...  The goal is to find new viewpoints or optimize sensor parameters for further measurements in order to classify an object and precisely locate its position.  ...  In [2] an appearance-based approach to active object recognition is proposed.  ... 
doi:10.1109/mfi.2008.4648083 dblp:conf/mfi/EidenbergerGFZ08 fatcat:tjage7sifngzvdnosiuuodgzom

Comparison of active SIFT-based 3D object recognition algorithms

Mogomotsi Keaikitse, Natasha Govender, Jonathan Warrell
2013 2013 Africon  
This paper compares two active object recognition systems. Both systems use SIFT features for object recognition, but use contrasting models, update and viewpoint selection strategies.  ...  The methods for integrating information across views used by the two systems are investigated. This is essential as this module is used to select the next best viewpoint.  ...  ACKNOWLEDGMENT This research was carried out with the financial support of the Council for Scientific and Industrial Research.  ... 
doi:10.1109/afrcon.2013.6757615 fatcat:lbs5vlhe4fhlxk7d5w7vwqkl6y

Active Random Forests: An Application to Autonomous Unfolding of Clothes [chapter]

Andreas Doumanoglou, Tae-Kyun Kim, Xiaowei Zhao, Sotiris Malassiotis
2014 Lecture Notes in Computer Science  
We present Active Random Forests, a novel framework to address active vision problems. State of the art focuses on best viewing parameters selection based on single view classifiers.  ...  The proposed framework is applied to the task of autonomously unfolding clothes by a robot, addressing the problem of best viewpoint selection in classification, grasp point and pose estimation of garments  ...  In the next Section we will describe how these objectives can be addressed using our Active Random Forests framework for efficient viewpoint selection.  ... 
doi:10.1007/978-3-319-10602-1_42 fatcat:vjsauv74cjct5efoqfbazvl4gy

Viewpoint detection models for sequential embodied object category recognition

David Meger, Ankur Gupta, James J Little
2010 2010 IEEE International Conference on Robotics and Automation  
This paper proposes a method for learning viewpoint detection models for object categories that facilitate sequential object category recognition and viewpoint planning.  ...  Simulation results verify that our viewpoint planning approach requires fewer viewpoints for confident recognition.  ...  This paper studies a method for Active Vision during category recognition.  ... 
doi:10.1109/robot.2010.5509703 dblp:conf/icra/MegerGL10 fatcat:xkeoqjzmbfdrjaby3urlzr34p4

Fast algorithms for large scale conditional 3D prediction

Liefeng Bo, Cristian Sminchisescu, Atul Kanaujia, Dimitris Metaxas
2008 2008 IEEE Conference on Computer Vision and Pattern Recognition  
The potential success of discriminative learning approaches to 3D reconstruction relies on the ability to efficiently train predictive algorithms using sufficiently many examples that are representative  ...  Recent research indicates that sparse conditional Bayesian Mixture of Experts (cMoE) models (e.g.  ...  Efficient forward selection methods for Gaussian Process learning are given in [25, 17] , for a tutorial see [8] .  ... 
doi:10.1109/cvpr.2008.4587578 dblp:conf/cvpr/BoSKM08 fatcat:qgwo6bbw2nhxlndf7ko2cfad4y

Active object recognition on a humanoid robot

Bjorn Browatzki, Vadim Tikhanoff, Giorgio Metta, Heinrich H. Bulthoff, Christian Wallraven
2012 2012 IEEE International Conference on Robotics and Automation  
Information is accumulated during the recognition process and used to select actions expected to be most beneficial in discriminating similar objects.  ...  Here, we propose a perception-driven, multisensory exploration and recognition scheme to actively resolve ambiguities that emerge at certain viewpoints.  ...  The authors wish to acknowledge Lorenzo Natale for his contributions to this work.  ... 
doi:10.1109/icra.2012.6225218 dblp:conf/icra/BrowatzkiTMBW12 fatcat:elfcifg3bbbrzoyxjux3mqaoci

Person Re-identification by Local Maximal Occurrence Representation and Metric Learning [article]

Shengcai Liao, Yang Hu, Xiangyu Zhu, Stan Z. Li
2015 arXiv   pre-print
An effective feature representation should be robust to illumination and viewpoint changes, and a discriminant metric should be learned to match various person images.  ...  We also present a practical computation method for XQDA, as well as its regularization.  ...  “Efficient model selection for regu- Computer Vision and Pattern Recognition, 2014. 6 larized linear discriminant analysis”. In Proceedings of the [29] Y. Liu, Y.  ... 
arXiv:1406.4216v2 fatcat:pbplpna2bfarhnanknsb3pjo4i

A Review on Video-Based Human Activity Recognition

Shian-Ru Ke, Hoang Thuc, Yong-Jin Lee, Jenq-Neng Hwang, Jang-Hee Yoo, Kyoung-Ho Choi
2013 Computers  
Three aspects for human activity recognition are addressed including core technology, human activity recognition systems, and applications from low-level to high-level representation.  ...  In the human activity recognition systems, three main types are mentioned, including single person activity recognition, multiple people interaction and crowd behavior, and abnormal activity recognition  ...  More specifically: Computers 2013, 2 120  The viewpoint issue remains the main challenge for human activity recognition.  ... 
doi:10.3390/computers2020088 fatcat:zb3wlmwjjvbfne2ck6uyjtffdq

Evidence filtering in a sequence of images for recognition

Sukhan Lee, Muhammad Ilyas, Kim Jaewoong, Ahmed Naguib
2012 2012 IEEE Applied Imagery Pattern Recognition Workshop (AIPR)  
We incorporate prior established Bayesian Evidence structure which embodied sufficient condition for recognition, to generate such interpretations.  ...  Furthermore, when robot moves, we do active recognition using particle filter framework in sequence of images to produce interpretation with highest weight and lowest error covariance.  ...  Bayesian Evidence structure is constructed for all these 16 viewpoint scenarios. A snapshot of Bayesian structure is given in table I and II.  ... 
doi:10.1109/aipr.2012.6528203 dblp:conf/aipr/LeeIKA12 fatcat:wbabhtdlcfew3cmtuhue6zw4sa

Active multi-view object recognition: A unifying view on online feature selection and view planning

Christian Potthast, Andreas Breitenmoser, Fei Sha, Gaurav S. Sukhatme
2016 Robotics and Autonomous Systems  
Adaptive action selection is a paradigm, offering great flexibil-*  ...  A strong desire exists to keep computation cost and energy consumption to a minimum when executing tasks like object recognition with a mobile robot.  ...  Acknowledgement The authors would like to thank Jörg Müller for his help on the development of the quadcopter robot platform.  ... 
doi:10.1016/j.robot.2016.06.013 fatcat:bxc5fmjyrnfytfg3hj7sugma6m

Conditional feature sensitivity: a unifying view on active recognition and feature selection

Xiang Sean Zhou, Comaniciu, Krishnan
2003 Proceedings Ninth IEEE International Conference on Computer Vision  
The objective of active recognition is to iteratively collect the next "best" measurements (e.g., camera angles or viewpoints), to maximally reduce ambiguities in recognition.  ...  Feature selection, on the other hand, focuses on the selection of a subset of measurements for a given classification task, but is not context sensitive (i.e., the decision does not depend on the current  ...  For example, most existing systems implicitly assume feature independence (which translates to viewpoint independence for object recognition using an active camera).  ... 
doi:10.1109/iccv.2003.1238668 dblp:conf/iccv/ZhouCK03 fatcat:pqyppjqljzafngt7zpy5u3glvu

Machine Learning Paradigms for Speech Recognition: An Overview

Li Deng, Xiao Li
2013 IEEE Transactions on Audio, Speech, and Language Processing  
Automatic Speech Recognition (ASR) has historically been a driving force behind many machine learning (ML) techniques, including the ubiquitously used hidden Markov model, discriminative learning, structured  ...  The paradigms presented and elaborated in this overview include: generative and discriminative learning; supervised, unsupervised, semi-supervised, and active learning; adaptive and multi-task learning  ...  Jeff Bilmes for contributions during the early phase (2010) of developing this paper, and for valuable discussions with Geoff Hinton, John Platt, Mark Gales, Nelson Morgan, Hynek Hermansky, Alex Acero,  ... 
doi:10.1109/tasl.2013.2244083 fatcat:fv4qulshnrh4fgzmzb45mkqwmq
« Previous Showing results 1 — 15 out of 3,536 results