Paper Title :Multi-Scale Fully Convolutional Neural Networks for Classification of Visual Objects

Author :Shu-Mei Lin, Hsueh-Fu Lu, Yuan-Hsiang Chang

Article Citation :Shu-Mei Lin ,Hsueh-Fu Lu ,Yuan-Hsiang Chang , (2018 ) " Multi-Scale Fully Convolutional Neural Networks for Classification of Visual Objects " , International Journal of Advances in Science, Engineering and Technology(IJASEAT) , pp. 1-5, Volume-6, Issue-4

Abstract : Recent studies have shown great potentials of the Convolutional Neural Networks (CNNs) to yield excellent results on visual classification tasks. While the CNNs could achieve translation-invariance by spatial convolution and pooling mechanisms, their ability to achieve scale-invariance is still limited. To overcome the challenge, we propose a multiscale fully CNNs network architecture that constitutes three types of multi-scale fusions, namely: (1) multi-size filters fusion; (2) multi-layer features fusion; and (3) multi-resolution I/Os fusion. Our CNNs’ architecture is designed to incorporate the fusions such that scale-invariance could be achieved. Using the CIFAR-10 and CIFAR-100 datasets as the benchmark for testing, our architecture has achieved classification accuracy with 96.6% (CIFAR-10) and 80.36% (CIFAR- 100), respectively. In conclusion, our multi-scale fully CNNs architecture has demonstrated the state-of-art classification performance based on published works to date. Index terms- Convolutional Neural Networks, CNN, multi-size kernel fusion, multi-layer feature fusion, multi-resolution I/O fusion.

Type : Research paper

Published : Volume-6, Issue-4

	|		PDF	|	Viewed - 71	|	Published on 2019-01-31