Dilated Residual Networks
Citations Over TimeTop 1% of 2017 papers
Abstract
Convolutional networks for image classification progressively reduce resolution until the image is represented by tiny feature maps in which the spatial structure of the scene is no longer discernible. Such loss of spatial acuity can limit image classification accuracy and complicate the transfer of the model to downstream applications that require detailed scene understanding. These problems can be alleviated by dilation, which increases the resolution of output feature maps without reducing the receptive field of individual neurons. We show that dilated residual networks (DRNs) outperform their non-dilated counterparts in image classification without increasing the models depth or complexity. We then study gridding artifacts introduced by dilation, develop an approach to removing these artifacts (degridding), and show that this further increases the performance of DRNs. In addition, we show that the accuracy advantage of DRNs is further magnified in downstream applications such as object localization and semantic segmentation.
Related Papers
- → ON THE ORGANIZATION OF RECEPTIVE FIELDS OF ORIENTATION-SELECTIVE UNITS RECORDED IN THE FISH TECTUM(2009)18 cited
- → Investigation into Sub-Receptive Fields of Retinal Ganglion Cells with Natural Images(2018)1 cited
- → Rabbit visual cortical neurons with simple and complex receptive fields(1978)1 cited
- → A model for activity-driven development of auditory receptive fields(1992)
- Investigation into Sub-Receptive Fields of Retinal Ganglion Cells with Natural Images(2018)