Distorted Representation Space Characterization Through Backpropagated Gradients
Authors:
- 1 September 2019
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 2651-2655
Abstract
In this paper, we utilize weight gradients from backpropagation to characterize the representation space learned by deep learning algorithms. We demonstrate the utility of such gradients in applications including perceptual image quality assessment and out-of-distribution classification. The applications are chosen to validate the effectiveness of gradients as features when the test image distribution is distorted from the train image distribution. In both applications, the proposed gradient based features outperform activation features. In image quality assessment, the proposed approach is compared with other state of the art approaches and is generally the top performing method on TID 2013 and MULTI-LIVE databases in terms of accuracy, consistency, linearity, and monotonic behavior. Finally, we analyze the effect of regularization on gradients using CURE-TSR dataset for out-of-distribution classification.Keywords
All Related Versions
This publication has 24 references indexed in Scilit:
- PerSIM: Multi-resolution image quality assessment in the perceptually uniform color domain2015
- Visual Domain Adaptation: A survey of recent advancesIEEE Signal Processing Magazine, 2015
- Image database TID2013: Peculiarities, results and perspectivesSignal Processing: Image Communication, 2015
- Representation Learning: A Review and New PerspectivesIEEE Transactions on Pattern Analysis and Machine Intelligence, 2013
- SR-SIM: A fast and high performance IQA index based on spectral residual2012
- No-Reference Image Quality Assessment in the Spatial DomainIEEE Transactions on Image Processing, 2012
- Regularization and Variable Selection Via the Elastic NetJournal of the Royal Statistical Society Series B: Statistical Methodology, 2005
- Multiscale structural similarity for image quality assessment2004
- Image quality assessment: from error visibility to structural similarityIEEE Transactions on Image Processing, 2004
- Learning representations by back-propagating errorsNature, 1986