Filters








9,050 Hits in 15.9 sec

Approximate Query Service on Autonomous IoT Cameras [article]

Mengwei Xu, Xiwen Zhang, Yunxin Liu, Xuanzhe Liu, Felix Xiaozhu Lin
2020 arXiv   pre-print
Can autonomous cameras do any useful analytics?  ...  Our response is iCam, a runtime for autonomous cameras, that is capable of producing useful analytics results with frugal camera resources. iCam is built atop two novel techniques: (1) For the first time  ...  Answering query with approximation and sampling Approximate query processing (AQP) [49] is to reduce the query delay on large-scale data.  ... 
arXiv:1909.00841v3 fatcat:qmncz4tjkbcfjoymxthjh2mz7m

A Survey of Performance Optimization in Neural Network-Based Video Analytics Systems [article]

Nada Ibrahim, Preeti Maurya, Omid Jafari, Parth Nagarkar
2021 arXiv   pre-print
Video analytics systems perform automatic events, movements, and actions recognition in a video and make it possible to execute queries on the video.  ...  As a result of a large number of video data that need to be processed, optimizing the performance of video analytics systems has become an important research topic.  ...  Depending on the query, it speeds up the query processing by at least two orders of magnitude.  ... 
arXiv:2105.14195v1 fatcat:klq3urgjsjhb7ef7qfdp6dly3u

2D visual place recognition for domestic service robots at night

James Mount, Michael Milford
2016 2016 IEEE International Conference on Robotics and Automation (ICRA)  
In a range of experiments over a domestic lawn and in a lounge room, we demonstrate that the proposed approach enables 2D localization at night, and analyse the effect on performance of varying odometry  ...  Finally we benchmark the new low light camera technology and show how it can enable robust place recognition even in an environment lit only by a moonless sky, raising the tantalizing possibility of being  ...  All processing was performed on a Windows 7 64-Bit machine running Matlab 2014. A.  ... 
doi:10.1109/icra.2016.7487686 dblp:conf/icra/MountM16 fatcat:jwfe7k4wbjg65gaudaklwb4rxi

2D Visual Place Recognition for Domestic Service Robots at Night [article]

James Mount, Michael Milford
2016 arXiv   pre-print
In a range of experiments over a domestic lawn and in a lounge room, we demonstrate that the proposed approach enables 2D localization at night, and analyse the effect on performance of varying odometry  ...  Finally we benchmark the new low light camera technology and show how it can enable robust place recognition even in an environment lit only by a moonless sky, raising the tantalizing possibility of being  ...  All processing was performed on a Windows 7 64-Bit machine running Matlab 2014. A.  ... 
arXiv:1605.07708v1 fatcat:kf3jts37gff3pco5nzieldyzr4

Monocular Camera Localization for Automated Vehicles Using Image Retrieval [article]

Eunhyek Joa, Francesco Borrelli
2021 arXiv   pre-print
We address the problem of finding the current position and heading angle of an autonomous vehicle in real-time using a single camera.  ...  We evaluate the proposed method using the KITTI odometry dataset and via closed-loop experiments with an indoor 1:10 autonomous vehicle.  ...  However, this method can only provide an approximate pose of a query image, not an exact pose.  ... 
arXiv:2109.06296v1 fatcat:tlu4443rangfhas5sdwggwpbaa

People Reidentification in a Camera Network

Icaro Oliveira de Oliveira, José Luiz de Souza Pio
2009 2009 Eighth IEEE International Conference on Dependable, Autonomic and Secure Computing  
The control component enables an automated communication of the query between the operator and the cameras. The query is processed from data supplied by the input of the operator.  ...  The reidentification or reacquisition problem consists essentially on the matching process of images acquired from different cameras. This work is applied in a monitored environment by cameras.  ...  This image contain data throught the selection of the camera where it shows the interest target. This process select one image. This image is sended to point detection and description process.  ... 
doi:10.1109/dasc.2009.33 dblp:conf/dasc/OliveiraP09 fatcat:mryuqtdiubc6jmvuynnyqd67si

Exploring Uncertainty in Conditional Multi-Modal Retrieval Systems [article]

Ahmed Taha, Yi-Ting Chen, Xitong Yang, Teruhisa Misu, Larry Davis
2019 arXiv   pre-print
Our approach is evaluated on two applications: person re-identification and autonomous car driving. Comparable state-of-the-art results are achieved on multiple datasets for the former application.  ...  This enables epistemic uncertainty estimation using dropout as a Bayesian approximation framework in retrieval. Accordingly, Monte Carlo (MC) sampling is leveraged to boost retrieval performance.  ...  An event retrieval evaluation using query-by-example is performed. Given a query event, similarity scores to all events are computed -i.e. a leave-one-out cross evaluation on the test split.  ... 
arXiv:1901.07702v1 fatcat:n4mfvhtvffgrfgageplfjk2tse

Leaving Flatland: Toward real-time 3D navigation

Benoit Morisset, Radu Bogdan Rusu, Aravind Sundaresan, Kris Hauser, Motilal Agrawal, Jean-Claude Latombe, Michael Beetz
2009 2009 IEEE International Conference on Robotics and Automation  
We report our first experiences with Leaving Flatland, an exploratory project that studies the key challenges of closing the loop between autonomous perception and action on challenging terrain.  ...  We test the map-building and motion-planning subsystems on real and synthetic data, and show that they have favorable computational performance for use in high-speed autonomous navigation.  ...  ACKNOWLEDGMENTS The authors thank Edward Van Reuth and DARPA for support on the "Leaving Flatland" project (contract #FA8650-04-C-7136).  ... 
doi:10.1109/robot.2009.5152715 dblp:conf/icra/MorissetRSHALB09 fatcat:gjhuksqsznc6reruddgfcjqd5e

Scene-Adaptive Human Detection with Incremental Active Learning

Ajay J. Joshi, Fatih Porikli
2010 2010 20th International Conference on Pattern Recognition  
For instance, a human detector trained with one set of images is unlikely to perform well in different scene conditions.  ...  For instance, a human detector trained with one set of images is unlikely to perform well in different scene conditions.  ...  The active selection process is usually iterative, wherein the algorithm queries the user for annotation on selected unlabeled examples, obtains user feedback, and appends the data to the training set.  ... 
doi:10.1109/icpr.2010.676 dblp:conf/icpr/JoshiP10 fatcat:gz5tfpoa6bhyjbl2l7urcflz34

Real-time image-based 6-DOF localization in large-scale environments

Hyon Lim, S. N. Sinha, M. F. Cohen, M. Uyttendaele
2012 2012 IEEE Conference on Computer Vision and Pattern Recognition  
Our algorithm runs at over 30 Hz on a laptop and at 12 Hz on a low-power, mobile computer suitable for onboard computation on a quadrotor micro aerial vehicle.  ...  From monocular video, our method continuously computes a precise 6-DOF camera pose, by efficiently tracking natural features and matching them to 3D points in the Sfm point cloud.  ...  For each descriptor q i , we perform a k-ANN query based on priority search using a kd-tree [3, 19] , which retrieves approximate nearest neighbors D i = {d ij } sorted by increasing distance {s ij },  ... 
doi:10.1109/cvpr.2012.6247782 dblp:conf/cvpr/LimSCU12 fatcat:ypbm2m6mwjd5jafv5vcqmigk4e

Autonomous Underwater Navigation and Optical Mapping in Unknown Natural Environments

Juan Hernández, Klemen Istenič, Nuno Gracias, Narcís Palomeras, Ricard Campos, Eduard Vidal, Rafael García, Marc Carreras
2016 Sensors  
We present an approach for navigating in unknown environments while, simultaneously, gathering information for inspecting underwater structures using an autonomous underwater vehicle (AUV).  ...  To accomplish this, we first use our pipeline for mapping and planning collision-free paths online, which endows an AUV with the capability to autonomously acquire optical data in close proximity.  ...  to use the Poisson method [74, 75] as one of the most representative approximation-based methods.  ... 
doi:10.3390/s16081174 pmid:27472337 pmcid:PMC5017340 fatcat:4mzz52lfyvfwriscz56w4cdbei

Repository of sensor data for autonomous driving research

Michael Shneier, Tommy Chang, Tsai Hong Hong, Geraldine S. Cheok, Harry Scott, Steven Legowik, Alan Lytle, Grant R. Gerhart, Charles M. Shoemaker, Douglas W. Gage
2003 Unmanned Ground Vehicle Technology V  
The sensors on the HMMWV include an imaging ladar, a color camera, color stereo, and inertial navigation (INS) and Global Positioning System (GPS).  ...  Data are collected using the NIST High Mobility Multi-purpose Wheeled Vehicle (HMMWV), an instrumented vehicle that can be driven manually or autonomously both on roads and off.  ...  More generally, the data are being collected and made available to researchers in order to increase the pace of research in sensor processing for autonomous vehicles.  ... 
doi:10.1117/12.485688 fatcat:vncyyjd6bbftzdwf5fuos36h7u

LONG-TERM VISUAL LOCALIZATION IN LARGE SCALE URBAN ENVIRONMENTS EXPLOITING STREET LEVEL IMAGERY

J. Meyer, D. Rettenmund, S. Nebiker
2020 ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences  
As the results on the RobotCar benchmark dataset are nearly on par with the top ranked approaches, we focused our investigations on reproducibility and performance with own data.  ...  The approximately 90% successfully oriented images of both datasets are a good indicator for the robustness of our approach.  ...  (SNSF) as part of EVAC project (No. 407540_167278) within the National Research Programme NFP75 on "Big Data".  ... 
doi:10.5194/isprs-annals-v-2-2020-57-2020 fatcat:eucqkbtdmrhdpcken3pwcxuaam

Autonomous Seabed Inspection for Environmental Monitoring [chapter]

Juan David Hernández, Klemen Istenic, Nuno Gracias, Rafael García, Pere Ridao, Marc Carreras
2015 Advances in Intelligent Systems and Computing  
We present an approach for navigating in unknown environments, while gathering information for inspecting underwater structures using an autonomous underwater vehicle (AUV).  ...  To accomplish this, we first use our framework for mapping and planning collision-free paths online, which endows an AUV with the capability to autonomously acquire optical data in close proximity.  ...  On that basis, the purpose of this paper is twofold: first, we build up on our previous work [4] a framework to endow an AUV with the capability to navigate autonomously in unknown environments while  ... 
doi:10.1007/978-3-319-27149-1_3 fatcat:obbpg2juu5alfiwvawht6nzml4

AMARO—An On-Board Ship Detection and Real-Time Information System

Katharina Willburger, Kurt Schwenk, Jörg Brauchle
2020 Sensors  
The AMARO (Autonomous Real-Time Detection of Moving Maritime Objects) project at DLR is a feasibility study of an on-board ship detection system involving on-board processing and real-time communication  ...  By processing and analyzing them on-board and transmitting the product data directly as ship position, heading, and velocity, the delay can be shortened to some minutes.  ...  We also thank Christian Mietner for building the AMARO-box and providing technical support, Sebastian Pless for organizing the flight campaign, Daniel Hein for providing software support for the MACS camera  ... 
doi:10.3390/s20051324 pmid:32121473 fatcat:32skajphyvcxner6vmdvllx2fe
« Previous Showing results 1 — 15 out of 9,050 results