Skip to main navigation Skip to search Skip to main content

Control+Shift: Generating Controllable Distribution Shifts

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

We propose a new method for generating realistic datasets with distribution shifts using any decoder-based generative model. Our approach systematically creates datasets with varying intensities of distribution shifts, facilitating a comprehensive analysis of model performance degradation. We then use these generated datasets to evaluate the performance of various commonly used networks and observe a consistent decline in performance with increasing shift intensity, even when the effect is almost perceptually unnoticeable to the human eye. We see this degradation even when using data augmentations. We also find that enlarging the training dataset beyond a certain point has no effect on the robustness and that stronger inductive biases increase robustness.

Original languageEnglish
Title of host publicationComputer Vision – ECCV 2024 Workshops, Proceedings
EditorsAlessio Del Bue, Cristian Canton, Jordi Pont-Tuset, Tatiana Tommasi
PublisherSpringer Science and Business Media Deutschland GmbH
Pages321-334
Number of pages14
ISBN (Print)9783031919060
DOIs
StatePublished - 2025
Event18th European Conference on Computer Vision, ECCV 2024 - Milan, Italy
Duration: 29 Sep 20244 Oct 2024

Publication series

NameLecture Notes in Computer Science
Volume15642 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference18th European Conference on Computer Vision, ECCV 2024
Country/TerritoryItaly
CityMilan
Period29/09/244/10/24

Bibliographical note

Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

Fingerprint

Dive into the research topics of 'Control+Shift: Generating Controllable Distribution Shifts'. Together they form a unique fingerprint.

Cite this