Wasserstein Distortion with Intrinsic $\sigma$-Maps

Abstract

Wasserstein distortion is a recently proposed family of distortion measures, controlled by a width parameter $\sigma$, that lifts fidelity and realism into a common framework. In previous implementations, calculating the Wasserstein distortion between two images relied on a companion saliency map or manual tuning to specify the width parameter $\sigma$ for each location in the image. We introduce a novel scheme for automatically generating an $\sigma$-map from the image itself.

Cite

Text

Qiu et al. "Wasserstein Distortion with Intrinsic $\sigma$-Maps." NeurIPS 2024 Workshops: Compression, 2024.

Markdown

[Qiu et al. "Wasserstein Distortion with Intrinsic $\sigma$-Maps." NeurIPS 2024 Workshops: Compression, 2024.](https://mlanthology.org/neuripsw/2024/qiu2024neuripsw-wasserstein/)

BibTeX

@inproceedings{qiu2024neuripsw-wasserstein,
  title     = {{Wasserstein Distortion with Intrinsic $\sigma$-Maps}},
  author    = {Qiu, Yang and Lin, Ziyuan and Wagner, Aaron B.},
  booktitle = {NeurIPS 2024 Workshops: Compression},
  year      = {2024},
  url       = {https://mlanthology.org/neuripsw/2024/qiu2024neuripsw-wasserstein/}
}