In response to the second question, we propose a Multi-Receptive-field Fusion Network (MRFNet) for multi-food recognition, which captures unique fine-grained features of Chinese food images using the multi-receptive-field pyramid network, fuses feature information from different receptive fields ...
The algorithm uses residual neural network (ResNet) to extract image features, uses recursive feature pyramid network (RFPN) to fuse features, and processes three outputs of RFPN by multiscale receptive field fusion (MRFF) to improve the ability of small target...
Finally, we propose a training strategy that applies different kernel sizes for feature maps of different scales so that the receptive field of the model can adapt to the scale changes of the feature maps to the greatest extent. The experiment on the NEU-DET dataset shows that our model ...
In the following parts, we will introduce the network from three aspects: (1) Multi-Feature Pyramid Module; (2) Receptive Field Block; (3) Double-Check Detection Network Module. Figure 2. Architecture of MFPNet. 3.1. Multi-Feature Pyramid Module Since the scale of objects in the ORSIs ...
Most spatial attentions operate on a linear structure, only accepting feature information from a fixed receptive field. This approach limits the model’s ability to process multi-scale spatial information. To address this, we integrate multi-scale spatial attention (MSA) and multi-scale channel ...
Next, we use a multiscale receptive field structure to help extract detailed feature information. To keep the size of the receptive field constant while reducing the inference time, we use a 3 × 3 dilated convolution kernel instead of a 5 × 5 convolution kernel and an asymmetric convolution...
We propose a parallel multi-scale receptive field framework that uses feature maps generated by the framework’s various depths. In addition, we also investigate the effect of the attention operation on crowd density prediction in a dynamic scene with objects of various scales. We propose an ensem...
(2023). MHLDet: A Multi-Scale and High-Precision Lightweight Object Detector Based on Large Receptive Field and Attention Mechanism for Remote Sensing Images. Remote Sensing, 15(18), 4625. https://doi.org/10.3390/rs15184625 Note that from the first issue of 2016, this journal uses article...