Boosting Object Detection with Zero-Shot Day-Night Domain Adaptation

TL;DR

This paper introduces a zero-shot domain adaptation method for object detection in low-light conditions, leveraging Retinex theory to improve generalization without requiring low-light training data.

Contribution

It proposes a novel Retinex-based reflectance learning and a redecomposition coherence strategy for zero-shot low-light object detection.

Findings

Strong generalization to low-light datasets

Effective without low-light training data

Outperforms existing methods in experiments

Abstract

Detecting objects in low-light scenarios presents a persistent challenge, as detectors trained on well-lit data exhibit significant performance degradation on low-light data due to low visibility. Previous methods mitigate this issue by exploring image enhancement or object detection techniques with real low-light image datasets. However, the progress is impeded by the inherent difficulties about collecting and annotating low-light images. To address this challenge, we propose to boost low-light object detection with zero-shot day-night domain adaptation, which aims to generalize a detector from well-lit scenarios to low-light ones without requiring real low-light data. Revisiting Retinex theory in the low-level vision, we first design a reflectance representation learning module to learn Retinex-based illumination invariance in images with a carefully designed illumination invariance reinforcement strategy. Next, an interchange-redecomposition-coherence procedure is introduced to improve over the vanilla Retinex image decomposition process by performing two sequential image decompositions and introducing a redecomposition cohering loss. Extensive experiments on ExDark, DARK FACE, and CODaN datasets show strong low-light generalizability of our method. Our code is available at https://github.com/ZPDu/DAI-Net.