Zero-Shot Day-Night Domain Adaptation with a Physics Prior

A. Lengyel, Sourav Garg, Michael Milford, J.C. van Gemert

Research output: Chapter in Book/Conference proceedings/Edited volumeConference contributionScientificpeer-review

45 Downloads (Pure)

Abstract

We explore the zero-shot setting for day-night domain adaptation. The traditional domain adaptation setting is to train on one domain and adapt to the target domain by exploiting unlabeled data samples from the test set. As gathering relevant test data is expensive and sometimes even impossible, we remove any reliance on test data imagery and instead exploit a visual inductive prior derived from physics-based reflection models for domain adaptation. We cast a number of color invariant edge detectors as trainable layers in a convolutional neural network and evaluate their robustness to illumination changes. We show that the color invariant layer reduces the day-night distribution shift in feature map activations throughout the network. We demonstrate improved performance for zero-shot day to night domain adaptation on both synthetic as well as natural datasets in various tasks, including classification, segmentation and place recognition.
Original languageEnglish
Title of host publicationProceedings of the IEEE/CVF International Conference on Computer Vision (ICCV)
EditorsL. O'Conner
Pages4399 - 4409
ISBN (Electronic)978-1-6654-0191-3
Publication statusPublished - 2021
Event2021 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW) - Virtual at Montreal, Canada
Duration: 11 Oct 202117 Oct 2021

Conference

Conference2021 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW)
Country/TerritoryCanada
CityVirtual at Montreal
Period11/10/2117/10/21

Fingerprint

Dive into the research topics of 'Zero-Shot Day-Night Domain Adaptation with a Physics Prior'. Together they form a unique fingerprint.

Cite this