How to scale hyperparameters for quickshift image segmentation

Damien Garreau
Proceedings of The 25th International Conference on Artificial Intelligence and Statistics, PMLR 151:5243-5275, 2022.

Abstract

Quickshift is a popular algorithm for image segmentation, used as a preprocessing step in many applications. Unfortunately, it is quite challenging to understand the hyperparameters’ influence on the number and shape of superpixels produced by the method. In this paper, we study theoretically a slightly modified version of the quickshift algorithm, with a particular emphasis on homogeneous image patches with i.i.d. pixel noise and sharp boundaries between such patches. Leveraging this analysis, we derive a simple heuristic to scale quickshift hyperparameters with respect to the image size, which we check empirically.

Cite this Paper

BibTeX
@InProceedings{pmlr-v151-garreau22a, title = { How to scale hyperparameters for quickshift image segmentation }, author = {Garreau, Damien}, booktitle = {Proceedings of The 25th International Conference on Artificial Intelligence and Statistics}, pages = {5243--5275}, year = {2022}, editor = {Camps-Valls, Gustau and Ruiz, Francisco J. R. and Valera, Isabel}, volume = {151}, series = {Proceedings of Machine Learning Research}, month = {28--30 Mar}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v151/garreau22a/garreau22a.pdf}, url = {https://proceedings.mlr.press/v151/garreau22a.html}, abstract = { Quickshift is a popular algorithm for image segmentation, used as a preprocessing step in many applications. Unfortunately, it is quite challenging to understand the hyperparameters’ influence on the number and shape of superpixels produced by the method. In this paper, we study theoretically a slightly modified version of the quickshift algorithm, with a particular emphasis on homogeneous image patches with i.i.d. pixel noise and sharp boundaries between such patches. Leveraging this analysis, we derive a simple heuristic to scale quickshift hyperparameters with respect to the image size, which we check empirically. } }
Endnote
%0 Conference Paper %T How to scale hyperparameters for quickshift image segmentation %A Damien Garreau %B Proceedings of The 25th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2022 %E Gustau Camps-Valls %E Francisco J. R. Ruiz %E Isabel Valera %F pmlr-v151-garreau22a %I PMLR %P 5243--5275 %U https://proceedings.mlr.press/v151/garreau22a.html %V 151 %X Quickshift is a popular algorithm for image segmentation, used as a preprocessing step in many applications. Unfortunately, it is quite challenging to understand the hyperparameters’ influence on the number and shape of superpixels produced by the method. In this paper, we study theoretically a slightly modified version of the quickshift algorithm, with a particular emphasis on homogeneous image patches with i.i.d. pixel noise and sharp boundaries between such patches. Leveraging this analysis, we derive a simple heuristic to scale quickshift hyperparameters with respect to the image size, which we check empirically.
APA
Garreau, D.. (2022). How to scale hyperparameters for quickshift image segmentation . Proceedings of The 25th International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 151:5243-5275 Available from https://proceedings.mlr.press/v151/garreau22a.html.

Related Material