Camera Bias in a Fine Grained Classification Task

07/16/2020
by   Philip T. Jackson, et al.
0

We show that correlations between the camera used to acquire an image and the class label of that image can be exploited by convolutional neural networks (CNN), resulting in a model that "cheats" at an image classification task by recognizing which camera took the image and inferring the class label from the camera. We show that models trained on a dataset with camera / label correlations do not generalize well to images in which those correlations are absent, nor to images from unencountered cameras. Furthermore, we investigate which visual features they are exploiting for camera recognition. Our experiments present evidence against the importance of global color statistics, lens deformation and chromatic aberration, and in favor of high frequency features, which may be introduced by image processing algorithms built into the cameras.

READ FULL TEXT

page 3

page 5

page 6

page 7

research
10/24/2019

Soft Prototyping Camera Designs for Car Detection Based on a Convolutional Neural Network

Imaging systems are increasingly used as input to convolutional neural n...
research
03/03/2016

First Steps Toward Camera Model Identification with Convolutional Neural Networks

Detecting the camera model used to shoot a picture enables to solve a wi...
research
09/29/2016

CNN Architectures for Large-Scale Audio Classification

Convolutional Neural Networks (CNNs) have proven very effective in image...
research
06/10/2021

The 2021 Hotel-ID to Combat Human Trafficking Competition Dataset

Hotel recognition is an important task for human trafficking investigati...
research
06/19/2021

Informative Class Activation Maps

We study how to evaluate the quantitative information content of a regio...
research
12/08/2019

Neural Network Generalization: The impact of camera parameters

We quantify the generalization of a convolutional neural network (CNN) t...
research
06/24/2018

Artwork Identification from Wearable Camera Images for Enhancing Experience of Museum Audiences

Recommendation systems based on image recognition could prove a vital to...

Please sign up or login with your details

Forgot password? Click here to reset