Search
Abstract
As modern convolutional neural networks become increasingly deeper, they also become slower and require high computational resources beyond the capabilities of many mobile and embedded platforms. To address this challenge, much of the recent research has focused on reducing the model size and computational complexity. In this paper, we propose a novel residual depth-separable convolution block, which is an improvement of the basic building block of MobileNet. We modified the original block by adding an identity shortcut connection (with zero-padding for increasing dimensions) from the input to the output. We demonstrated that the modified architecture with the width multiplier (α) set to 0.92 slightly outperforms the accuracy and inference time of the baseline MobileNet (α = 1) on the challenging Places365 dataset while reducing the number of parameters by 14%.
Citations
-
2
CrossRef
-
0
Web of Science
-
1
Scopus
Authors (4)
Cite as
Full text
full text is not available in portal
Keywords
Details
- Category:
- Conference activity
- Type:
- publikacja w wydawnictwie zbiorowym recenzowanym (także w materiałach konferencyjnych)
- Title of issue:
- New Trends in Databases and Information Systems strony 315 - 324
- Language:
- English
- Publication year:
- 2019
- Bibliographic description:
- Brzeski A., Grinholc K., Nowodworski K., Przybyłek A.: Residual MobileNets// New Trends in Databases and Information Systems/ : , 2019, s.315-324
- DOI:
- Digital Object Identifier (open in new tab) 10.1007/978-3-030-30278-8_33
- Verified by:
- Gdańsk University of Technology
seen 152 times
Recommended for you
Column base fixity in steel moment frames: Observations from instrumented buildings
- T. Falborski,
- A. Hassan,
- A. Kanvinde