Global Pooling, More than Meets the Eye: Position Information is Encoded Channel-Wise in CNNs

08/17/2021
by   Md Amirul Islam, et al.
28

In this paper, we challenge the common assumption that collapsing the spatial dimensions of a 3D (spatial-channel) tensor in a convolutional neural network (CNN) into a vector via global pooling removes all spatial information. Specifically, we demonstrate that positional information is encoded based on the ordering of the channel dimensions, while semantic information is largely not. Following this demonstration, we show the real world impact of these findings by applying them to two applications. First, we propose a simple yet effective data augmentation strategy and loss function which improves the translation invariance of a CNN's output. Second, we propose a method to efficiently determine which channels in the latent representation are responsible for (i) encoding overall position information or (ii) region-specific positions. We first show that semantic segmentation has a significant reliance on the overall position channels to make predictions. We then show for the first time that it is possible to perform a `region-specific' attack, and degrade a network's performance in a particular part of the input. We believe our findings and demonstrated applications will benefit research areas concerned with understanding the characteristics of CNNs.

READ FULL TEXT
POST COMMENT

Comments

There are no comments yet.

Authors

page 3

page 5

page 8

01/28/2021

Position, Padding and Predictions: A Deeper Look at Position Information in CNNs

In contrast to fully connected networks, Convolutional Neural Networks (...
02/12/2018

Object Detection with Mask-based Feature Encoding

Region-based Convolutional Neural Networks (R-CNNs) have achieved great ...
01/27/2021

Shape or Texture: Understanding Discriminative Features in CNNs

Contrasting the previous evidence that neurons in the later layers of a ...
12/08/2021

Fully Attentional Network for Semantic Segmentation

Recent non-local self-attention methods have proven to be effective in c...
07/24/2017

Traffic scene recognition based on deep cnn and vlad spatial pyramids

Traffic scene recognition is an important and challenging issue in Intel...
01/13/2021

Convolutional Neural Nets: Foundations, Computations, and New Applications

We review mathematical foundations of convolutional neural nets (CNNs) w...
07/18/2018

Location Augmentation for CNN

CNNs have made a tremendous impact on the field of computer vision in th...

Code Repositories

This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.