Training and Testing Texture Similarity Metrics for Structurally Lossless Compression
- 15 February 2024
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 33 (10577149) , 1614-1626
Abstract
We present a systematic approach for training and testing structural texture similarity metrics (STSIMs) so that they can be used to exploit texture redundancy for structurally lossless image compression. The training and testing is based on a set of image distortions that reflect the characteristics of the perturbations present in natural texture images. We conduct empirical studies to determine the perceived similarity scale across all pairs of original and distorted textures. We then introduce a data-driven approach for training the Mahalanobis formulation of STSIM based on the resulting annotated texture pairs. Experimental results demonstrate that training results in significant improvements in metric performance. We also show that the performance of the trained STSIM metrics is competitive with state of the art metrics based on convolutional neural networks, at substantially lower computational cost.Keywords
This publication has 32 references indexed in Scilit:
- PieAPP: Perceptual Image-Error Assessment Through Pairwise Preference2018
- The Unreasonable Effectiveness of Deep Features as a Perceptual Metric2018
- Image Style Transfer Using Convolutional Neural Networks2016
- Improved side matching for matched-texture coding2014
- On the assessment of the quality of textures in visual media2010
- Perceptual Criteria for Image Quality Evaluation2005
- A Parametric Texture Model Based on Joint Statistics of Complex Wavelet CoefficientsInternational Journal of Computer Vision, 2000
- Perceptual quality metrics applied to still image compressionSignal Processing, 1998
- Shiftable multiscale transformsIEEE Transactions on Information Theory, 1992
- A law of comparative judgment.Psychological Review, 1927