Cross-Domain Weakly-Supervised Object Detection through Progressive Domain Adaptation
This addresses the challenge of object detection in varied domains with limited annotation, but it is incremental as it builds on existing weakly-supervised and domain adaptation methods.
The paper tackles the problem of detecting objects across different image domains without instance-level annotations in the target domain, achieving improvements of 5 to 20 percentage points in mAP compared to baselines.
Can we detect common objects in a variety of image domains without instance-level annotations? In this paper, we present a framework for a novel task, cross-domain weakly supervised object detection, which addresses this question. For this paper, we have access to images with instance-level annotations in a source domain (e.g., natural image) and images with image-level annotations in a target domain (e.g., watercolor). In addition, the classes to be detected in the target domain are all or a subset of those in the source domain. Starting from a fully supervised object detector, which is pre-trained on the source domain, we propose a two-step progressive domain adaptation technique by fine-tuning the detector on two types of artificially and automatically generated samples. We test our methods on our newly collected datasets containing three image domains, and achieve an improvement of approximately 5 to 20 percentage points in terms of mean average precision (mAP) compared to the best-performing baselines.