Where-What Network 5: Dealing with scales for objects in complex backgrounds

Author

Song, Xiaoying ; Zhang, Wenqiang ; Weng, Juyang

Author_Institution

Sch. of Comput. Sci., Fudan Univ., Shanghai, China

fYear

2011

fDate

July 31 2011-Aug. 5 2011

Firstpage

2795

Lastpage

2802

Abstract

The biologically-inspired developmental Where-What Networks (WWN) are general purpose visuomotor networks for detecting and recognizing objects from complex backgrounds, modeling the dorsal and ventral streams of the biological visual cortex. The networks are designed for the attention and recognition problem. The architecture in previous versions were meant for a single scale of foreground. This paper focuses on Where-What Network-5 (WWN-5), the extension for multiple scales. WWN-5 can learn three concepts of an object: type, location and scale.

Keywords

learning (artificial intelligence); neural nets; object detection; object recognition; visual perception; attention problem; biological visual cortex; biologically-inspired developmental where-what network; complex background; dorsal stream modeling; general purpose visuomotor network; learning; neural network; object detection; object location; object recognition; object scale; object type; recognition problem; ventral stream modeling; Computational modeling; Computer architecture; Educational institutions; Neurons; Retina; Shape; attention; neural networks; object recognition; recognition; scale invariance;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks (IJCNN), The 2011 International Joint Conference on

Conference_Location

San Jose, CA

ISSN

2161-4393

Print_ISBN

978-1-4244-9635-8

Type

conf

DOI

10.1109/IJCNN.2011.6033587

Filename

6033587