Filters








9,244 Hits in 8.6 sec

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  
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.  ...  This paper presents an exclusively parametric approach for the state estimation and decision making process to achieve very low computational complexity and short calculation times.  ...  CONCLUSION This paper describes a fast method for state estimation and viewpoint selection, which is suited for high dimensional state spaces and entirely computes the probability distributions and uncertainty  ... 
doi:10.1109/mfi.2008.4648083 dblp:conf/mfi/EidenbergerGFZ08 fatcat:tjage7sifngzvdnosiuuodgzom

3D-RCNN: Instance-Level 3D Object Reconstruction via Render-and-Compare

Abhijit Kundu, Yin Li, James M. Rehg
2018 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition  
We present a fast inverse-graphics framework for instance-level 3D scene understanding.  ...  We train a deep convolutional network that learns to map image regions to the full 3D shape and pose of all object instances in the image.  ...  Apart from evaluating on joint detection and pose estimation task, we also provide controlled study for 3D pose estimation with fixed object detection input.  ... 
doi:10.1109/cvpr.2018.00375 dblp:conf/cvpr/KunduLR18 fatcat:vl7mt2prfzdjxgwo5gs4pkil4y

3D Hand Pose Detection in Egocentric RGB-D Images [chapter]

Grégory Rogez, Maryam Khademi, J. S. Supančič III, J. M. M. Montiel, Deva Ramanan
2015 Lecture Notes in Computer Science  
We propose to use task and viewpoint specific synthetic training exemplars in a discriminative detection framework. We also exploit the depth features for a sparser and faster detection.  ...  We focus on the task of hand pose estimation from egocentric viewpoints.  ...  To do so efficiently, we have exploited a simple depth cue for fast detection.  ... 
doi:10.1007/978-3-319-16178-5_25 fatcat:wloilrcfxrbfbeitmjfr74hna4

A Survey on Model Based Approaches for 2D and 3D Visual Human Pose Recovery

Xavier Perez-Sala, Sergio Escalera, Cecilio Angulo, Jordi Gonzàlez
2014 Sensors  
In this paper, we define a general taxonomy to group model based approaches for Human Pose Recovery, which is composed of five main modules: appearance, viewpoint, spatial relations, temporal consistence  ...  Human Pose Recovery has been studied in the field of Computer Vision for the last 40 years.  ...  Pictorial structures are a general framework for object detection widely used for people detection and human pose estimation [23, 54] .  ... 
doi:10.3390/s140304189 pmid:24594613 pmcid:PMC4003939 fatcat:m37lzyl5tne6jgu3aurjz6stiy

HUMAN ACTION POSELETS ESTIMATION VIA COLOR G-SURF IN STILL IMAGES

M. Favorskaya, D. Novikov, Y. Savitskaya
2015 The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences  
The proposed approach was tested on samples from PASCAL Visual Object Classes (VOC) Dataset and Challenge 2010 providing accuracy 61-68% for all possible 3D poses locations and 82-86% for front poses locations  ...  Human activity is a persistent subject of interest in the last decade. On the one hand, video sequences provide a huge volume of motion information in order to recognize the human active actions.  ...  CONCLUSION In this study, the poselets estimation is proposed for recognition of human active actions.  ... 
doi:10.5194/isprsarchives-xl-5-w6-51-2015 fatcat:j5tflaktljcetfwjpyodygeiku

Probabilistic occlusion estimation in cluttered environments for active perception planning

Robert Eidenberger, Raoul Zoellner, Josef Scharinger
2009 2009 IEEE/ASME International Conference on Advanced Intelligent Mechatronics  
This work also deals with the probabilistic computation of object occlusions for probabilistic action planning in order to select the most profitable prospective viewpoint.  ...  It tackles the problems of representing detection and transition uncertainties in multi-object scenes without knowledge of the total number of objects in the scenario.  ...  RELATED WORK The research in active perception brought up very sophisticated approaches for viewpoint evaluation and next best view planning [3][4] [5] [6] .  ... 
doi:10.1109/aim.2009.5229779 fatcat:i47objekpng2laq2danvaan53q

Beyond Standard Benchmarks: Parameterizing Performance Evaluation in Visual Object Tracking

Luka Cehovin Zajc, Alan Lukezic, Ales Leonardis, Matej Kristan
2017 2017 IEEE International Conference on Computer Vision (ICCV)  
[9] address the virtual worlds realism levels, ambient parametrization learning and performance evaluation, however, only for vehicle detection.  ...  [18] propose online virtual worlds creation for drone tracking evaluation, but using only a single type of the object, without motion parametrization, produces a low level of realism. Vig et al.  ... 
doi:10.1109/iccv.2017.360 dblp:conf/iccv/ZajcLLK17 fatcat:64ikihvhr5dbzbuzhho3hwuqiy

Beyond standard benchmarks: Parameterizing performance evaluation in visual object tracking [article]

Luka Čehovin Zajc, Alan Lukežič, Aleš Leonardis, Matej Kristan
2017 arXiv   pre-print
Object-to-camera motion produces a variety of apparent motion patterns that significantly affect performance of short-term visual trackers.  ...  We have created an evaluation system, constructed a fully annotated dataset of omnidirectional videos and the generators for typical motion patterns.  ...  [10] address the virtual worlds realism levels, ambient parametrization learning and performance evaluation, however, only for vehicle detection.  ... 
arXiv:1612.00089v2 fatcat:qpkavw7bxrhgfem4fn2dmymr3m

ICCV Program

2007 2007 IEEE 11th International Conference on Computer Vision  
Konrad Schindler, Luc Van Gool: [PDF paper] Coupled Detection and Trajectory Estimation for Multi-Object Tracking Bi Song, Amit K.  ...  and the Estimation of Pose and Shape Bo Wu, Nevatia Ram: Cluster Boosted Tree Classifier for Multi-View, Multi-Pose Object Detection [PDF paper] [PDF paper] Video Henrik Malm, Magnus Oskarsson  ... 
doi:10.1109/iccv.2007.4408821 fatcat:csxpst65sbagxhouhz5yuj3qbu

Deep Reinforcement Learning for Active Human Pose Estimation [article]

Erik Gärtner, Aleksis Pirinen, Cristian Sminchisescu
2020 arXiv   pre-print
In extensive experiments with the Panoptic multi-view setup, and for complex scenes containing multiple people, we show that our model learns to select viewpoints that yield significantly more accurate  ...  In this paper we address the problem of an active observer with freedom to move and explore the scene spatially -- in 'time-freeze' mode -- and/or temporally, by selecting informative viewpoints that improve  ...  Also, the estimates of already detected people improve as they get fused from multiple viewpoints.  ... 
arXiv:2001.02024v2 fatcat:nmqdjxmmm5aq3abzmli3bdtogq

Deep Reinforcement Learning for Active Human Pose Estimation

Erik Gärtner, Aleksis Pirinen, Cristian Sminchisescu
2020 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
In extensive experiments with the Panoptic multi-view setup, and for complex scenes containing multiple people, we show that our model learns to select viewpoints that yield significantly more accurate  ...  In this paper we address the problem of an active observer with freedom to move and explore the scene spatially – in 'time-freeze' mode – and/or temporally, by selecting informative viewpoints that improve  ...  Also, the estimates of already detected people improve as they get fused from multiple viewpoints.  ... 
doi:10.1609/aaai.v34i07.6714 fatcat:j32l3gxotffpbkgvngurnqcr5u

Volumetric segmentation of brain images using parallel genetic algorithms

Yong Fan, Tianzi Jiang, D.J. Evans
2002 IEEE Transactions on Medical Imaging  
We demonstrate that the method successfully overcomes numerical instability and is capable of generating an accurate and robust anatomic descriptor for complex objects in the human brain, such as the lateral  ...  First, an objective function is defined.  ...  ACKNOWLEDGMENT The authors are highly grateful to the anonymous referees for their significant and constructive critiques and suggestions, which improve the paper very much.  ... 
doi:10.1109/tmi.2002.803126 pmid:12472263 fatcat:crrampfvnjallpgrjr4ijdg7em

Viewpoint Invariant Matching via Developable Surfaces [chapter]

Bernhard Zeisl, Kevin Köser, Marc Pollefeys
2012 Lecture Notes in Computer Science  
We address the topic of wide baseline matching between two RGBD images, i.e. finding correspondences from largely different viewpoints for recognition, model fusion or loop detection.  ...  This allows viewpoint invariant matching also in scenes with curved architectural elements or with objects like bottles, cans or (partial) cones and others.  ...  Note, that for the bottom experiment depth estimates are noisy and contain a considerable amount of errors, leading to degraded parameter estimation for the detected cone.  ... 
doi:10.1007/978-3-642-33868-7_7 fatcat:rac4z3kinfcvtkhjnuvt2ux7g4

Familiar Face Detection in 180ms

Matteo Visconti di Oleggio Castello, M. Ida Gobbini, Adrian G Dyer
2015 PLoS ONE  
The visual system is tuned for rapid detection of faces, with the fastest choice saccade to a face at 100ms.  ...  Saccades to faces with object distractors were even faster (110 to 120 ms) and equivalent for familiar and unfamiliar faces, indicating that familiarity does not affect ultra-rapid saccades.  ...  Acknowledgments The authors would like to thank Alireza Soltani for help with the eye tracker, Sébastien Crouzet for sharing the original stimuli of his experiment for a pilot used to validate the set  ... 
doi:10.1371/journal.pone.0136548 pmid:26305788 pmcid:PMC4549263 fatcat:3n5ubvc3zvedzapyli67bgbn5i

Nonmyopic View Planning for Active Object Classification and Pose Estimation

Nikolay Atanasov, Bharath Sankaran, Jerome Le Ny, George J. Pappas, Kostas Daniilidis
2014 IEEE Transactions on robotics  
One of the central problems in computer vision is the detection of semantically important objects and the estimation of their pose.  ...  This paper proposes an active approach to object detection in which the point of view of a mobile depth camera is controlled.  ...  Cohen for their help with ROS and the PR2.  ... 
doi:10.1109/tro.2014.2320795 fatcat:woeobpojd5h2bkyqwy4dz7dgaa
« Previous Showing results 1 — 15 out of 9,244 results