We propose a novel and flexible anchor mechanism named MetaAnchor for object
detection frameworks. Unlike many previous detectors model anchors via a
predefined manner, in MetaAnchor anchor functions could be dynamically
generated from the arbitrary customized prior boxes. Taking advantage of weight
prediction, MetaAnchor is able to work with most of the anchor-based object
detection systems such as RetinaNet. Compared with the predefined anchor
scheme, we empirically find that MetaAnchor is more robust to anchor settings
and bounding box distributions; in addition, it also shows the potential on
transfer tasks. Our experiment on COCO detection task shows that MetaAnchor
consistently outperforms the counterparts in various scenarios.
Yang, Tong, Zhang, Xiangyu, Zhang, Wenqiang, Sun, Jian
We propose a novel and flexible anchor mechanism named MetaAnchor for object detection frameworks. Unlike many previous detectors model anchors via a predefined manner, in MetaAnchor anchor functions could be dynamically generated from the arbitrary customized prior boxes. Taking advantage of weight prediction, MetaAnchor is able to work with most of the anchor-based object detection systems such as RetinaNet. Compared with the predefined anchor scheme, we empirically find that MetaAnchor is more robust to anchor settings and bounding box distributions; in addition, it also shows the potential on transfer tasks. Our experiment on COCO detection task shows that MetaAnchor consistently outperforms the counterparts in various scenarios.
https://arxiv.org/abs/1807.00980