graphs learned convolutional neural networks features graph CNN public pedestrian detection datasets multireceptive field graph convolutional neural networks general object detection deep learning pedestrian detection task pedestrian detection performance multireceptive field-based framework single-shot pedestrian ...
Multi Receptive Field Network for Semantic Segmentationdoi:10.1109/WACV45572.2020.9093264Yuan JianlongZelu DengWang ShuLuo ZhenboIEEE
He, Z., Cao, Y., Du, L., Xu, B., Yang, J., Cao, Y., Tang, S., Zhuang, Y.: MRFN: multi-receptive-field network for fast and accurate single image super-resolution. IEEE Trans. Multimed. 22(4), 1042–1054 (2020). https://doi.org/10.1109/TMM.2019.2937688 Article Google Sch...
Such direction-dependent receptive field organization was observed in every cortical layer. We conclude that the spatial structure of receptive fields in the barrel cortex is not an intrinsic property of the neuron but depends on the properties of sensory input. 展开 ...
为了解决上述问题,我们设计了一种多尺度扩张残差块(MDRB)fMDRB multi-scale dilated residual block (MDRB),它不仅可以有效地扩大感受野 receptive field 以感知帧之间的大像素运动, 还可以 在扩张卷积的帮助下可以很好地保留对象边界细节 捕获多尺度上下文信息。
因此提出 Direct multi-hop Attention based Graph neural Network (DAGN),在注意力机制中加入多跳信息,从邻居节点扩展到非邻居节点,增加每一层网络中的感受野(receptive field)。同时,DAGCN采用diffusion prior的方法来计算节点对上的attention values。实验结果表明能取得 state-of-the-art 的结果。 Introduction 目前...
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、...
N. Receptive fields of single neurones in the cat’s striate cortex. J. Physiol. 148, 574–591 (1959). Article CAS PubMed PubMed Central Google Scholar Chichilnisky, E. J. A simple white noise analysis of neuronal light responses. Netw. Comput. Neural Syst. 12, 199–213 (2001). ...
Each logit judges the plausibility of a segment of the input that corresponds to its receptive field. We refer interested readers to [26] for more architectural details. 对于鉴别器,我们使用与 [26] 相同的多分辨率卷积架构。 三个结构相同的鉴别器应用于不同分辨率的输入音频: 原始、2x 下采样和 4x...
时域卷积网络(TCN),由扩展的一维卷积(dilated 1-D convolutions)组成,比起其他模型可以用更少的参数构建大的时域感受野(temporal receptive field) TCN的应用:[18]用于文字转语音(text-to speech),[19]-[23]用于语音增强,[24]用于语音分离。 文献[20]构建的语音增强系统使用了多分支的TCN(MB-TCN),有效地进行...