Shape Adaptor: A Learnable Resizing Module

08/03/2020
by   Shikun Liu, et al.
12

We present a novel resizing module for neural networks: shape adaptor, a drop-in enhancement built on top of traditional resizing layers, such as pooling, bilinear sampling, and strided convolution. Whilst traditional resizing layers have fixed and deterministic reshaping factors, our module allows for a learnable reshaping factor. Our implementation enables shape adaptors to be trained end-to-end without any additional supervision, through which network architectures can be optimised for each individual task, in a fully automated way. We performed experiments across seven image classification datasets, and results show that by simply using a set of our shape adaptors instead of the original resizing layers, performance increases consistently over human-designed networks, across all datasets. Additionally, we show the effectiveness of shape adaptors on two other applications: network compression and transfer learning. The source code is available at: https://github.com/lorenmt/shape-adaptor.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
06/07/2018

Efficient semantic image segmentation with superpixel pooling

In this work, we evaluate the use of superpixel pooling layers in deep n...
research
06/01/2023

Dilated Convolution with Learnable Spacings: beyond bilinear interpolation

Dilated Convolution with Learnable Spacings (DCLS) is a recently propose...
research
03/19/2022

Representation-Agnostic Shape Fields

3D shape analysis has been widely explored in the era of deep learning. ...
research
12/07/2021

Dilated convolution with learnable spacings

Dilated convolution is basically a convolution with a wider kernel creat...
research
03/19/2023

Partial Network Cloning

In this paper, we study a novel task that enables partial knowledge tran...
research
12/07/2021

CapsProm: A Capsule Network For Promoter Prediction

Locating the promoter region in DNA sequences is of paramount importance...
research
09/03/2019

Image Inpainting with Learnable Bidirectional Attention Maps

Most convolutional network (CNN)-based inpainting methods adopt standard...

Please sign up or login with your details

Forgot password? Click here to reset