Self-Supervision & Meta-Learning for One-Shot Unsupervised Cross-Domain Detection

TL;DR

This paper introduces a novel one-shot unsupervised cross-domain object detection method that leverages self-supervision and meta-learning to adapt to new domains using only a single target sample, addressing practical scenarios like social media feeds.

Contribution

It proposes a multi-task architecture with self-supervised meta-training for one-shot domain adaptation in object detection, suitable for real-world applications with limited target data.

Findings

Outperforms recent cross-domain detection methods on social media datasets.

Effective one-shot adaptation using only a single target sample.

Demonstrates robustness in unseen domain scenarios.

Abstract

Deep detection approaches are powerful in controlled conditions, but appear brittle and fail when source models are used off-the-shelf on unseen domains. Most of the existing works on domain adaptation simplify the setting and access jointly both a large source dataset and a sizable amount of target samples. However this scenario is unrealistic in many practical cases as when monitoring image feeds from social media: only a pretrained source model is available and every target image uploaded by the users belongs to a different domain not foreseen during training. We address this challenging setting by presenting an object detection algorithm able to exploit a pre-trained source model and perform unsupervised adaptation by using only one target sample seen at test time. Our multi-task architecture includes a self-supervised branch that we exploit to meta-train the whole model with single-sample cross-domain episodes, and prepare to the test condition. At deployment time the self-supervised task is iteratively solved on any incoming sample to one-shot adapt on it. We introduce a new dataset of social media image feeds and present a thorough benchmark with the most recent cross-domain detection methods showing the advantages of our approach.