Zero Cost Improvements for General Object Detection Network

To solve the contradiction of increasing computational cost along with the precision improvement in modern object detection networks, it is necessary to research precision improvement without extra cost. In this work, two modules are proposed to improve detection precision with zero cost, which are focus on FPN and detection head improvement for general object detection networks. The scale attention mechanism is employed to efficiently fuse multi-level feature maps with less parameters, which is called SA-FPN module. For the sake of the correlation between classification head and regression head, sequential head is used to take the place of widely-used parallel head, which is called Seq-HEAD module. To evaluate the effectiveness, the two modules are applied to some modern state-of-art object detection networks, including anchor-based and anchor-free. Experiment results on coco dataset show that the networks with the two modules can surpass original networks by 1.1 AP and 0.8 AP with zero cost for anchor-based and anchor-free networks, respectively.