Log In Sign Up

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

by   Md Amirul Islam, et al.

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.


page 3

page 5

page 8


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

In contrast to fully connected networks, Convolutional Neural Networks (...

DPANET:Dual Pooling Attention Network for Semantic Segmentation

Image segmentation is a historic and significant computer vision task. W...

Object Detection with Mask-based Feature Encoding

Region-based Convolutional Neural Networks (R-CNNs) have achieved great ...

Shape or Texture: Understanding Discriminative Features in CNNs

Contrasting the previous evidence that neurons in the later layers of a ...

Fully Attentional Network for Semantic Segmentation

Recent non-local self-attention methods have proven to be effective in c...

Volumetric Transformer Networks

Existing techniques to encode spatial invariance within deep convolution...

Convolutional Neural Nets: Foundations, Computations, and New Applications

We review mathematical foundations of convolutional neural nets (CNNs) w...

Code Repositories