-
1. 【论文解读】SKNet网络（自适应调整感受野尺寸） - 知乎
Link: https://zhuanlan.zhihu.com/p/80513438
Description: WebSelective Kernel Networks论文： Selective Kernel Networks 代码：implus/SKNet 摘要在标准卷积神经网络（CNN）中，每层中人工神经元的感受野被设计成共享相同的大小。在神经科学界众所周知，视觉皮层神经元的感…
-
2. SKNet解读-CSDN博客
Link: https://blog.csdn.net/luxinfeng666/article/details/102070894
Description: WebOct 4, 2019 · SKNet的全称是“Selective Kernel Network”,它和SENet都是一个团队提出来的，SENet是对通道执行注意力机制，而SKNet则是对卷积核执行注意力机制，即让网络自己选择合适的卷积核。
-
3. 【CV中的Attention机制】SKNet-SENet的提升版 - 知乎
Link: https://zhuanlan.zhihu.com/p/102034839
Description: Web前言：SKNet是SENet的加强版，是attention机制中的与SE同等地位的一个模块，可以方便地添加到现有的网络模型中，对分类问题，分割问题有一定的提升。. 1. SKNet. SKNet是SENet的加强版，结合了SE opetator、Merge-and-Run Mappings以及attention on inception block的产物。其最 …
-
4. SKNet: Selective Kernel Networks - GitHub
Link: https://github.com/implus/SKNet
Description: WebWe are retraining a model and it will come soon in 2-3 days. 20190326_Update: SKNet-101 model is ready. Attention weights correspond to object scales in low/middle layers We look deep into the selection distributions from the perspective of classes on SK_2_3 (low), SK_3_4 (middle), SK_5_3 (high) layers:
-
5. [1903.06586] Selective Kernel Networks - arXiv.org
Link: https://arxiv.org/abs/1903.06586
Description: WebMar 15, 2019 · Multiple SK units are stacked to a deep network termed Selective Kernel Networks (SKNets). On the ImageNet and CIFAR benchmarks, we empirically show that SKNet outperforms the existing state-of-the-art architectures with lower model complexity.
-
6. 【CV中的Attention机制】基于SENet的改进-SKNet - 知乎
Link: https://zhuanlan.zhihu.com/p/142228301
Description: Web代码地址： https:// github.com/implus/SKNet; Selective Kernel Networks（SKNet）发表在CVPR 2019，SKNet是SENet的加强版，结合了SE opetator、Merge-and-Run Mappings以及attention on inception block的产物。名为SK模块, 可以自适应调节自身的感受野。
-
7. Selective Kernel Networks | Papers With Code
Link: https://paperswithcode.com/paper/selective-kernel-networks
Description: WebMultiple SK units are stacked to a deep network termed Selective Kernel Networks (SKNets). On the ImageNet and CIFAR benchmarks, we empirically show that SKNet outperforms the existing state-of-the-art architectures with lower model complexity. Detailed analyses show that the neurons in SKNet can capture target objects with different scales ...
-
8. Selective Kernel Networks | IEEE Conference Publication | IEEE …
Link: https://ieeexplore.ieee.org/document/8954149
Description: WebJun 20, 2019 · Multiple SK units are stacked to a deep network termed Selective Kernel Networks (SKNets). On the ImageNet and CIFAR benchmarks, we empirically show that SKNet outperforms the existing state-of-the-art architectures with lower model complexity.
-
9. SKNet Explained | Papers With Code
Link: https://paperswithcode.com/method/sknet
Description: WebSKNet is a type of convolutional neural network that employs selective kernel units, with selective kernel convolutions, in its architecture. This allows for a type of attention where the network can learn to attend to different receptive fields.
-
10. Selective Kernel Networks（SKNet）解析，附Pytorch实现代码 …
Link: https://blog.csdn.net/ITOMG/article/details/89673593
Description: WebNov 17, 2020 · 由多个SK单元组成SKNet，SKNet中的神经元能够捕获不同尺度的目标物体。. 这个网络主要分为 Split，Fuse，Select 三个操作。. Split算子产生多条不同核大小的路径，上图中的模型只设计了两个不同大小的卷积. 最低0.47元/天 解锁文章. 文章浏览阅读9.4k次，点赞7次 ...
