文献类型：期刊文献

英文题名：DRPN: Making CNN Dynamically Handle Scale Variation

作者：Peng, Jingchao[1]; Zhao, Haitao[1]; Hu, Zhengwei[1]; Zhao, Kaijie[1]; Wang, Zhongze[1]

机构：[1] East China University of Science and Technology, Automation Department, School of Information Science and Engineering, China

年份：2021

外文期刊名：arXiv

收录：EI(收录号：20210419611)

语种：英文

外文关键词：Convolution - Large dataset - Parameterization

摘要：Based on our observations of infrared targets, serious scale variation along within sequence frames has high-frequently occurred. In this paper, we propose a dynamic re-parameterization network (DRPN) to deal with the scale variation and balance the detection precision between small targets and large targets in infrared datasets. DRPN adopts the multiple branches with different sizes of convolution kernels and the dynamic convolution strategy. Multiple branches with different sizes of convolution kernels have different sizes of receptive fields. Dynamic convolution strategy makes DRPN adaptively weight multiple branches. DRPN can dynamically adjust the receptive field according to the scale variation of the target. Besides, in order to maintain effective inference in the test phase, the multi-branch structure is further converted to a single-branch structure via the re-parameterization technique after training. Extensive experiments on FLIR, KAIST, and InfraPlane datasets demonstrate the effectiveness of our proposed DRPN. The experimental results show that detectors using the proposed DRPN as the basic structure rather than SKNet or TridentNet obtained the best performances. Copyright ? 2021, The Authors. All rights reserved.