Recent Studies Around the Neocognitron

Shouno, Hayaru

doi:10.1007/978-3-540-69158-7_109

Cited by 5 publications

(3 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Comparison with the conventional convolution-net, such like Neocognitron, Le-Net and Lowe model [6], [7], [8], [10], the most difference points is description manner. The convolution-net often apply Gabor function like filter in order to extract line/edge segment in the image, and sub-sampling, which is called blurring, is carried out for local pattern deformation.…”

Section: Conclusion and Discussionmentioning

confidence: 99%

“…Thus, Hubel & Wiesel proposed a kind of hierarchy between these cells, that is, a complex cell may gather the outputs of simple cells that respond to same preferred stimulus but have slightly different receptive fields. The convolution-net has a hierarchy of these types of cells, and connects this hierarchy alternately [6], [7], [10]. Figure 2 is a schematic diagram of the convolution-net proposed by Mutch & Lowe [10].…”

Section: Summary Of a Convolution-net Of Mutch And Lowementioning

confidence: 99%

“…Such hierarchical image processing models, which are inspired from the brain, have been proposed. Fukushima proposed "Neocognitron", and applied it to handwritten character recognition [6], [7]. LeCun also proposed "Le-Net" series and showed good performance in the meaning of the recognition accuracy [8].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Hierarchical Extension of the HOG Model Implemented in the Convolution-net for Human Detection

Arakaki

Shouno

Takahashi

et al. 2012

IPSJ Online Transactions

Self Cite

View full text Add to dashboard Cite

For the detection of generic objects in the field of image processing, histograms of orientation gradients (HOG) is discussed for these years. The performance of the classification system using HOG shows a good result. However, the performance of using HOG descriptor would be influenced by the detecting object size. In order to overcome this problem, we introduce a kind of hierarchy inspired from the convolution-net, which is a model of our visual processing system in the brain. The hierarchical HOG (H-HOG) integrates several scales of HOG descriptors in its architecture, and represents the input image as the combinatorial of more complex features rather than that of the orientation gradients. We investigate the H-HOG performance and compare with the conventional HOG. In the result, we obtain the better performance rather than the conventional HOG. Especially the size of representation dimension is much smaller than the conventional HOG without reducing the detecting performance.

show abstract