Fig. 1From: An improved object detection algorithm based on multi-scaled and deformable convolutional neural networksArchitecture of the multi-scaled deformable convolutional neural network framework. The framework mainly consists of two components. (1) The backbone network based on Darknet53, which includes residual structures and deformable convolution. (2) The object detection network, based on multi-scaled detection and feature fusionBack to article page