Feature pyramidReceptive fieldObject detection datasetScale variation is one of the challenges in object detection. In this paper, we design a Multi-Level Feature Fusion Pyramid Network (MLFFPN) that can fuse features with different receptive fields so as to produce reliable object representations ...
The DCN module can use its adaptive receptive field to focus on the area of interest for calculation and play a role in texture feature enhancement. Finally, the perceptual loss is chosen as the regularization item of the loss function, which makes style features of the restored image closer ...
Pyramid Pooling (PP) increases the scale of the receptive field, thus effectively integrating multi-scale features. Inspired by the structure of the UNet++ (Zhou et al., 2018), proposed a multi-side output fusion network (UNet++_MSOF) that combines multi-level change maps with improved ...
In this paper, we propose a model for UAV detection called DoubleM-Net, which contains multi-scale spatial pyramid pooling-fast (MS-SPPF) and Multi-Path Adaptive Feature Pyramid Network (MPA-FPN). DoubleM-Net utilizes the MS-SPPF module to extract feature maps of multiple receptive field ...
3.2 Feature encoder module Due to the limited receptive field and the shallow depth of U-Net networks, it is insufficient for effective multiscale feature extraction. To address this issue, the U-Net architecture is replaced with the pre-trained ResNet-34 [41] in the feature encoder block to...
Although FPN-like feature fusion models have achieved remarkable results in the field of computer vision, they still have some shortcomings. On the one hand, as mentioned in the paper33, in the pyramid feature fusion structure, the deep feature information is transferred to the shallow features ...
We integrate the Style Normalization and Restitution (SNR) module for domain generalization, Receptive Field Blocks (RFBs) for fine-grained detail capture, and a twin-branch Global Context Module (TBGCM) for multiscale context information. We enhance lateral connections within the Feature Pyramid ...
能构建更深的网络,增大“receptive field” 模糊图像和清晰图像在数值上本身就比较相近,因此仅仅让网络学习两者的差异也够了 整体网络结构 文中选择了K=3的“multi-scale architecture”,输入、输出的“Gaussian pyramid patches”大小为{256×256,128×128,64×64}。 B_{k},L_{k},S_{k} 分别表示模糊图像、...
Specifically, we present a consecutive multiscale feature-learning network (CMSFL-Net) that employs a consecutive feature-learning approach based on the usage of various feature maps with different receptive fields to achieve faster training/inference and higher accuracy. In the conducted experiments ...
Firstly, Feature Pyramid Network (FPN) is used to combine the multiple feature layers of Convolutional Neural Network (CNN) to predict the geometric properties of text, which can be used to expand the receptive field of each pixel and thus help to detect more large-scale texts. Secondly, a ...