Cross-Domain Weakly-Supervised Object Detection through Progressive Domain Adaptation

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.

PDF Abstract CVPR 2018 PDF CVPR 2018 Abstract


Introduced in the Paper:

Clipart1k Watercolor2k Comic2k

Used in the Paper:


Results from the Paper

Ranked #3 on Weakly Supervised Object Detection on Comic2k (using extra training data)

     Get a GitHub badge
Task Dataset Model Metric Name Metric Value Global Rank Uses Extra
Training Data
Result Benchmark
Weakly Supervised Object Detection Clipart1k DT+PL MAP 46.0 # 4
Weakly Supervised Object Detection Comic2k DT+PL (+extra) MAP 42.2 # 3
Weakly Supervised Object Detection Comic2k DT+PL MAP 37.2 # 4
Weakly Supervised Object Detection Watercolor2k DT+PL (+extra) MAP 59.1 # 3
Weakly Supervised Object Detection Watercolor2k DT+PL MAP 54.3 # 7


No methods listed for this paper. Add relevant methods here