papers AI Learner
The Github is limit! Click to go to the new site.

Feature Evaluation of Deep Convolutional Neural Networks for Object Recognition and Detection

2015-09-25
Hirokatsu Kataoka, Kenji Iwata, Yutaka Satoh

Abstract

In this paper, we evaluate convolutional neural network (CNN) features using the AlexNet architecture and very deep convolutional network (VGGNet) architecture. To date, most CNN researchers have employed the last layers before output, which were extracted from the fully connected feature layers. However, since it is unlikely that feature representation effectiveness is dependent on the problem, this study evaluates additional convolutional layers that are adjacent to fully connected layers, in addition to executing simple tuning for feature concatenation (e.g., layer 3 + layer 5 + layer 7) and transformation, using tools such as principal component analysis. In our experiments, we carried out detection and classification tasks using the Caltech 101 and Daimler Pedestrian Benchmark Datasets.

Abstract (translated by Google)
URL

https://arxiv.org/abs/1509.07627

PDF

https://arxiv.org/pdf/1509.07627


Similar Posts

Comments