Better object detection when domain change with balanced domains using BD-DIR
Better object detection when domain change with balanced domains using BD-DIR
Domain-Invariant Proposals based on a Balanced Domain Classifier for Object Detection
arXiv paper abstract https://arxiv.org/abs/2202.05941v1
arXiv PDF paper https://arxiv.org/pdf/2202.05941v1.pdf
Object recognition ... from research on deep learning ... the performance ... improved significantly, especially when training and test data are drawn from similar distributions.
However, mismatching distributions, i.e., domain shifts, lead to a significant performance drop.
... build domain-invariant detectors by learning domain classifiers via adversarial training.
... mitigate the domain shift further by introducing a domain adaptation component at the region level within Faster R-CNN.
... embed a domain classification network in the region proposal network~(RPN) using adversarial learning.
... demonstrate the effectiveness and robustness of ... object detection approach in domain shift scenarios.
Please like and share this post if you enjoyed it using the buttons at the bottom!
Stay up to date. Subscribe to my posts https://morrislee1234.wixsite.com/website/contact
Web site with my other posts by category https://morrislee1234.wixsite.com/website
Comments