# On Translation Invariance in CNNs: Convolutional Layers can Exploit Absolute Spatial Location

Source: [On Translation Invariance in CNNs: Convolutional Layers can Exploit Absolute Spatial Location](<https://www.sota2.com/research/paper/68fa67ae1e209131cea17bda>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-25T22:20:05.971Z

Authors: Osman Semih Kayhan, Jan C. van Gemert

Published: 2020-03-16T08:00:06.000Z

## Abstract

In this paper we challenge the common assumption that convolutional layers in modern CNNs are translation invariant. We show that CNNs can and will exploit the absolute spatial location by learning filters that respond exclusively to particular absolute locations by exploiting image boundary effects. Because modern CNNs filters have a huge receptive field, these boundary effects operate even far from the image boundary, allowing the network to exploit absolute spatial location all over the image. We give a simple solution to remove spatial location encoding which improves translation invariance and thus gives a stronger visual inductive bias which particularly benefits small data sets. We broadly demonstrate these benefits on several architectures and various applications such as image classification, patch matching, and two video classification datasets.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2003.07064>)

[PDF](<https://arxiv.org/pdf/2003.07064.pdf>)