Characterization of the Distortion-Perception Tradeoff for Finite Channels with Arbitrary Metrics
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909092752130048 |
|---|---|
| author | Freirich, Dror Weinberger, Nir Meir, Ron |
| author_facet | Freirich, Dror Weinberger, Nir Meir, Ron |
| contents | Whenever inspected by humans, reconstructed signals should not be distinguished from real ones. Typically, such a high perceptual quality comes at the price of high reconstruction error, and vice versa. We study this distortion-perception (DP) tradeoff over finite-alphabet channels, for the Wasserstein-$1$ distance induced by a general metric as the perception index, and an arbitrary distortion matrix. Under this setting, we show that computing the DP function and the optimal reconstructions is equivalent to solving a set of linear programming problems. We provide a structural characterization of the DP tradeoff, where the DP function is piecewise linear in the perception index. We further derive a closed-form expression for the case of binary sources. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_02265 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Characterization of the Distortion-Perception Tradeoff for Finite Channels with Arbitrary Metrics Freirich, Dror Weinberger, Nir Meir, Ron Information Theory Signal Processing Machine Learning Whenever inspected by humans, reconstructed signals should not be distinguished from real ones. Typically, such a high perceptual quality comes at the price of high reconstruction error, and vice versa. We study this distortion-perception (DP) tradeoff over finite-alphabet channels, for the Wasserstein-$1$ distance induced by a general metric as the perception index, and an arbitrary distortion matrix. Under this setting, we show that computing the DP function and the optimal reconstructions is equivalent to solving a set of linear programming problems. We provide a structural characterization of the DP tradeoff, where the DP function is piecewise linear in the perception index. We further derive a closed-form expression for the case of binary sources. |
| title | Characterization of the Distortion-Perception Tradeoff for Finite Channels with Arbitrary Metrics |
| topic | Information Theory Signal Processing Machine Learning |
| url | https://arxiv.org/abs/2402.02265 |