Shape Adaptor: A Learnable Resizing Module
This work addresses the limitation of non-adaptive resizing in neural networks for computer vision tasks, offering an incremental enhancement to existing methods.
The paper tackles the problem of fixed reshaping factors in neural network resizing layers by introducing a learnable resizing module called shape adaptor, which consistently improves performance across seven image classification datasets and shows effectiveness in network compression and transfer learning.
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.