Dynamic Feature Pyramid Networks for Object Detection
This work provides a more efficient FPN for object detection frameworks, offering a better trade-off between accuracy and computational cost for practitioners and researchers in computer vision.
This paper addresses the computational burden of Feature Pyramid Networks (FPNs) in object detection by proposing a dynamic FPN (DyFPN). The DyFPN adaptively selects branches for feature calculation, achieving a 40% reduction in FLOPs compared to inception FPN while maintaining similar high performance on the MS-COCO benchmark.
Feature pyramid network (FPN) is a critical component in modern object detection frameworks. The performance gain in most of the existing FPN variants is mainly attributed to the increase of computational burden. An attempt to enhance the FPN is enriching the spatial information by expanding the receptive fields, which is promising to largely improve the detection accuracy. In this paper, we first investigate how expanding the receptive fields affect the accuracy and computational costs of FPN. We explore a baseline model called inception FPN in which each lateral connection contains convolution filters with different kernel sizes. Moreover, we point out that not all objects need such a complicated calculation and propose a new dynamic FPN (DyFPN). The output features of DyFPN will be calculated by using the adaptively selected branch according to a dynamic gating operation. Therefore, the proposed method can provide a more efficient dynamic inference for achieving a better trade-off between accuracy and computational cost. Extensive experiments conducted on MS-COCO benchmark demonstrate that the proposed DyFPN significantly improves performance with the optimal allocation of computation resources. For instance, replacing inception FPN with DyFPN reduces about 40% of its FLOPs while maintaining similar high performance.