Lightweight Active Object Retrieval with Weak Classifiers
In the last few years, there has been a steadily growing interest in autonomous vehicles and robotic systems. While many of these agents are expected to have limited resources, these systems should be able to dynamically interact with other objects in their environment. We present an approach where...
Main Authors: | , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2018-03-01
|
Series: | Sensors |
Subjects: | |
Online Access: | http://www.mdpi.com/1424-8220/18/3/801 |
_version_ | 1811304429230489600 |
---|---|
author | László Czúni Metwally Rashad |
author_facet | László Czúni Metwally Rashad |
author_sort | László Czúni |
collection | DOAJ |
description | In the last few years, there has been a steadily growing interest in autonomous vehicles and robotic systems. While many of these agents are expected to have limited resources, these systems should be able to dynamically interact with other objects in their environment. We present an approach where lightweight sensory and processing techniques, requiring very limited memory and processing power, can be successfully applied to the task of object retrieval using sensors of different modalities. We use the Hough framework to fuse optical and orientation information of the different views of the objects. In the presented spatio-temporal perception technique, we apply active vision, where, based on the analysis of initial measurements, the direction of the next view is determined to increase the hit-rate of retrieval. The performance of the proposed methods is shown on three datasets loaded with heavy noise. |
first_indexed | 2024-04-13T08:07:05Z |
format | Article |
id | doaj.art-89b7646e25cb44a2924032bd8cc0267f |
institution | Directory Open Access Journal |
issn | 1424-8220 |
language | English |
last_indexed | 2024-04-13T08:07:05Z |
publishDate | 2018-03-01 |
publisher | MDPI AG |
record_format | Article |
series | Sensors |
spelling | doaj.art-89b7646e25cb44a2924032bd8cc0267f2022-12-22T02:55:08ZengMDPI AGSensors1424-82202018-03-0118380110.3390/s18030801s18030801Lightweight Active Object Retrieval with Weak ClassifiersLászló Czúni0Metwally Rashad1Department of Electrical Engineering and Information Systems, University of Pannonia, Veszprém 8200, HungaryDepartment of Electrical Engineering and Information Systems, University of Pannonia, Veszprém 8200, HungaryIn the last few years, there has been a steadily growing interest in autonomous vehicles and robotic systems. While many of these agents are expected to have limited resources, these systems should be able to dynamically interact with other objects in their environment. We present an approach where lightweight sensory and processing techniques, requiring very limited memory and processing power, can be successfully applied to the task of object retrieval using sensors of different modalities. We use the Hough framework to fuse optical and orientation information of the different views of the objects. In the presented spatio-temporal perception technique, we apply active vision, where, based on the analysis of initial measurements, the direction of the next view is determined to increase the hit-rate of retrieval. The performance of the proposed methods is shown on three datasets loaded with heavy noise.http://www.mdpi.com/1424-8220/18/3/801object retrievalHough transformationsensor fusionactive vision |
spellingShingle | László Czúni Metwally Rashad Lightweight Active Object Retrieval with Weak Classifiers Sensors object retrieval Hough transformation sensor fusion active vision |
title | Lightweight Active Object Retrieval with Weak Classifiers |
title_full | Lightweight Active Object Retrieval with Weak Classifiers |
title_fullStr | Lightweight Active Object Retrieval with Weak Classifiers |
title_full_unstemmed | Lightweight Active Object Retrieval with Weak Classifiers |
title_short | Lightweight Active Object Retrieval with Weak Classifiers |
title_sort | lightweight active object retrieval with weak classifiers |
topic | object retrieval Hough transformation sensor fusion active vision |
url | http://www.mdpi.com/1424-8220/18/3/801 |
work_keys_str_mv | AT laszloczuni lightweightactiveobjectretrievalwithweakclassifiers AT metwallyrashad lightweightactiveobjectretrievalwithweakclassifiers |