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

TL;DR

This paper introduces a zero-shot domain adaptation method for day-night scenarios using physics-based reflection priors, improving robustness without relying on test data imagery.

Contribution

It proposes a novel zero-shot domain adaptation approach leveraging physics-inspired priors and color invariant features, eliminating the need for test data during adaptation.

Findings

Color invariant layers reduce day-night distribution shift.

Improved zero-shot adaptation performance on synthetic and natural datasets.

Effective across classification, segmentation, and place recognition tasks.

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.