Camera Bias in a Fine Grained Classification Task
| Autor: | Philip T. Jackson, Boguslaw Obara, Christopher J. Holder, Ning Jia, Jon Stonehouse, Stephen Bonner |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2020 |
| Předmět: |
FOS: Computer and information sciences
Contextual image classification Computer science business.industry Distortion (optics) Computer Vision and Pattern Recognition (cs.CV) ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Computer Science - Computer Vision and Pattern Recognition Image segmentation Convolutional neural network law.invention Lens (optics) law Pattern recognition (psychology) Digital image processing Chromatic aberration Computer vision Artificial intelligence business |
| Zdroj: | IJCNN |
| Popis: | 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. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|