However, previous attention approaches typically are of a single receptive field, which are difficult to capture rich structural affinities from different scales, harming salient region inferring effect. In this paper, we propose a multi-receptive field attention (MRFA) for person re-identification. ...
In this paper, our work introduces the Multi-receptive Field Distillation Network (MFD-Net) as a comprehensive solution to address challenges in seismic velocity modeling. The integration of the SR Transformer Block as a decoder enhances the network’s ability to capture long-range dependencies, the...
To tackle these issues, in this paper, we propose a novel multi-receptive field attention (MRFA) module that utilizes filters of various sizes to help network focusing on informative pixels. Besides, we present a view-specific mechanism that guides attention module to handle the variation of ...
The fusion performance of the mentioned deep learning-based fusion methods outperforms most of traditional-based methods in information mining and timeliness, but it still exists some inadequacies: 1) The constant size of the convolution kernel in dense network makes the receptive field unchanged, ...
Empirical findings demonstrate that a fully trained MRF-PINN proficiently reconstructs electromagnetic field distributions within complex nanomaterials within a mere tens of milliseconds of inference time. Such quasi real-time capabilities herald a novel approach to supplant the arduous forward calculation ...
A large receptive field is suitable for large object detection, and a small receptive field is beneficial for small object detection. However, the detector receptive field does not match the object. The bounding box offset has greatly influenced small object detection. Therefore, it is difficult to...
To improve the model’s feature extraction ability for targets, a receptive field enhancement module based on dilated convolution and shared weights is proposed. By expanding the receptive field of the feature map, it can extract the detailed features and local information of multi-scale targets. ...
In this paper, a new computational model of self-organised Hebbian learning (SOHL) is proposed to work on a multi-resolution image pyramid for the problem of visual receptive field learning. Receptive fields of both orientation and spatial frequency selectivity are observed in the authors' ...
Multi-receptive Field Distillation Network for seismic velocity model building Engineering Applications of Artificial Intelligence Volume 133, Part F, July 2024, Page 108547 Purchase options CorporateFor R&D professionals working in corporate organizations. Academic and personalFor academic or personal use onl...
In this paper, we introduce a novel multi-scale receptive field GAT. To learn the semantic structure relationships of land covers and reduce the computational complexity, an STM module is proposed, in which a 1D CNN is designed to suppress the noise of original HSI datasets and extract the sp...