344 0

Incremental Learning from a Single Seed Image for Object Detection

Title
Incremental Learning from a Single Seed Image for Object Detection
Author
임종우
Keywords
Databases; Feature extraction; Three-dimensional displays; Object detection; Computational modeling; Visualization; Solid modeling
Issue Date
2015-09
Publisher
IROS
Citation
Intelligent Robots and Systems (IROS), 2015 IEEE/RSJ International Conference on, Page. 1905-1912
Abstract
In this paper, we propose a novel online multiobject learning and detection algorithm. From single seed images of the target objects, our algorithm detects these objects in the input sequence, and incrementally updates the databases with the detection results. Reasonably sized databases are maintained as graphs of the registered images, while new views of the objects are added as the detection proceeds. The importance of the registered images is computed using our ranking algorithm, and redundant images are pruned from the database. The proposed method fully utilizes graphical representation to detect and recognize objects. A 3D model of a candidate object is built on-the-fly using the retrieved images, and initially undetected features are hallucinated for further matching and verification. This process improves the detection performance compared to the baseline algorithm. Object/background feature classification and object-likelihood maps effectively keep noisy background features from being added to the databases. The experimental results demonstrate that the proposed algorithm efficiently maintains the object databases and achieves better performance.
URI
http://ieeexplore.ieee.org/document/7353627/http://hdl.handle.net/20.500.11754/27377
ISBN
978-1-4799-9994-1; 978-1-4799-9993-4
DOI
10.1109/IROS.2015.7353627
Appears in Collections:
COLLEGE OF ENGINEERING[S](공과대학) > COMPUTER SCIENCE AND ENGINEERING(컴퓨터공학부) > Articles
Files in This Item:
There are no files associated with this item.
Export
RIS (EndNote)
XLS (Excel)
XML


qrcode

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

BROWSE