diff --git a/docs/zh_cn/models/models_intro.md b/docs/zh_cn/models/models_intro.md
index d195e160b6f7705a5c0e79a59d665e9b0d43397d..ea7f86b1505d51d8942ab5585fbc03cb4fd2a014 100644
--- a/docs/zh_cn/models/models_intro.md
+++ b/docs/zh_cn/models/models_intro.md
@@ -8,13 +8,13 @@
![](../../images/models/mobile_arm_top1.png)
- ResNet及其Vd系列
- - ResNet系列`[1]`([论文地址](http://openaccess.thecvf.com/content_cvpr_2016/html/He_Deep_Residual_Learning_CVPR_2016_paper.html))
+ - ResNet系列[[1](#ref1)]([论文地址](http://openaccess.thecvf.com/content_cvpr_2016/html/He_Deep_Residual_Learning_CVPR_2016_paper.html))
- [ResNet18](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet18_pretrained.tar)
- [ResNet34](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet34_pretrained.tar)
- [ResNet50](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_pretrained.tar)
- [ResNet101](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet101_pretrained.tar)
- [ResNet152](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet152_pretrained.tar)
- - ResNet_vc、ResNet_vd系列([论文地址](https://arxiv.org/abs/1812.01187))
+ - ResNet_vc、ResNet_vd系列[[2](#ref2)]([论文地址](https://arxiv.org/abs/1812.01187))
- [ResNet50_vc](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_vc_pretrained.tar)
- [ResNet18_vd](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet18_vd_pretrained.tar)
- [ResNet34_vd](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet34_vd_pretrained.tar)
@@ -27,7 +27,7 @@
- 移动端系列
- - MobileNetV3系列([论文地址](https://arxiv.org/abs/1905.02244))
+ - MobileNetV3系列[[3](#ref3)]([论文地址](https://arxiv.org/abs/1905.02244))
- [MobileNetV3_large_x0_35](https://paddle-imagenet-models-name.bj.bcebos.com/MobileNetV3_large_x0_35_pretrained.tar)
- [MobileNetV3_large_x0_5](https://paddle-imagenet-models-name.bj.bcebos.com/MobileNetV3_large_x0_5_pretrained.tar)
- [MobileNetV3_large_x0_75](https://paddle-imagenet-models-name.bj.bcebos.com/MobileNetV3_large_x0_75_pretrained.tar)
@@ -41,7 +41,7 @@
- [MobileNetV3_large_x1_0_ssld](https://paddle-imagenet-models-name.bj.bcebos.com/MobileNetV3_large_x1_0_ssld_pretrained.tar)
- [MobileNetV3_large_x1_0_ssld_int8](https://paddle-imagenet-models-name.bj.bcebos.com/MobileNetV3_large_x1_0_ssld_int8_pretrained.tar)
- [MobileNetV3_small_x1_0_ssld](https://paddle-imagenet-models-name.bj.bcebos.com/MobileNetV3_small_x1_0_ssld_pretrained.tar)
- - MobileNetV2系列([论文地址](https://arxiv.org/abs/1801.04381))
+ - MobileNetV2系列[[4](#ref4)]([论文地址](https://arxiv.org/abs/1801.04381))
- [MobileNetV2_x0_25](https://paddle-imagenet-models-name.bj.bcebos.com/MobileNetV2_x0_25_pretrained.tar)
- [MobileNetV2_x0_5](https://paddle-imagenet-models-name.bj.bcebos.com/MobileNetV2_x0_5_pretrained.tar)
- [MobileNetV2_x0_75](https://paddle-imagenet-models-name.bj.bcebos.com/MobileNetV2_x0_75_pretrained.tar)
@@ -49,13 +49,13 @@
- [MobileNetV2_x1_5](https://paddle-imagenet-models-name.bj.bcebos.com/MobileNetV2_x1_5_pretrained.tar)
- [MobileNetV2_x2_0](https://paddle-imagenet-models-name.bj.bcebos.com/MobileNetV2_x2_0_pretrained.tar)
- [MobileNetV2_ssld](https://paddle-imagenet-models-name.bj.bcebos.com/MobileNetV2_ssld_pretrained.tar)
- - MobileNetV1系列([论文地址](https://arxiv.org/abs/1704.04861))
+ - MobileNetV1系列[[5](#ref5)]([论文地址](https://arxiv.org/abs/1704.04861))
- [MobileNetV1_x0_25](https://paddle-imagenet-models-name.bj.bcebos.com/MobileNetV1_x0_25_pretrained.tar)
- [MobileNetV1_x0_5](https://paddle-imagenet-models-name.bj.bcebos.com/MobileNetV1_x0_5_pretrained.tar)
- [MobileNetV1_x0_75](https://paddle-imagenet-models-name.bj.bcebos.com/MobileNetV1_x0_75_pretrained.tar)
- [MobileNetV1](https://paddle-imagenet-models-name.bj.bcebos.com/MobileNetV1_pretrained.tar)
- [MobileNetV1_ssld](https://paddle-imagenet-models-name.bj.bcebos.com/MobileNetV1_ssld_pretrained.tar)
- - ShuffleNetV2系列([论文地址](https://arxiv.org/abs/1807.11164))
+ - ShuffleNetV2系列[[6](#ref6)]([论文地址](https://arxiv.org/abs/1807.11164))
- [ShuffleNetV2_x0_25](https://paddle-imagenet-models-name.bj.bcebos.com/ShuffleNetV2_x0_25_pretrained.tar)
- [ShuffleNetV2_x0_33](https://paddle-imagenet-models-name.bj.bcebos.com/ShuffleNetV2_x0_33_pretrained.tar)
- [ShuffleNetV2_x0_5](https://paddle-imagenet-models-name.bj.bcebos.com/ShuffleNetV2_x0_5_pretrained.tar)
@@ -66,7 +66,7 @@
- SEResNeXt与Res2Net系列
- - ResNeXt系列([论文地址](https://arxiv.org/abs/1611.05431))
+ - ResNeXt系列[[7](#ref7)]([论文地址](https://arxiv.org/abs/1611.05431))
- [ResNeXt50_32x4d](https://paddle-imagenet-models-name.bj.bcebos.com/ResNeXt50_32x4d_pretrained.tar)
- [ResNeXt50_64x4d](https://paddle-imagenet-models-name.bj.bcebos.com/ResNeXt50_64x4d_pretrained.tar)
- [ResNeXt101_32x4d](https://paddle-imagenet-models-name.bj.bcebos.com/ResNeXt101_32x4d_pretrained.tar)
@@ -80,7 +80,7 @@
- [ResNeXt101_vd_64x4d](https://paddle-imagenet-models-name.bj.bcebos.com/ResNeXt101_vd_64x4d_pretrained.tar)
- [ResNeXt152_vd_32x4d](https://paddle-imagenet-models-name.bj.bcebos.com/ResNeXt152_vd_32x4d_pretrained.tar)
- [ResNeXt152_vd_64x4d](https://paddle-imagenet-models-name.bj.bcebos.com/ResNeXt152_vd_64x4d_pretrained.tar)
- - SE_ResNet_vd系列([论文地址](https://arxiv.org/abs/1709.01507))
+ - SE_ResNet_vd系列[[8](#ref8)]([论文地址](https://arxiv.org/abs/1709.01507))
- [SE_ResNet18_vd](https://paddle-imagenet-models-name.bj.bcebos.com/SE_ResNet18_vd_pretrained.tar)
- [SE_ResNet34_vd](https://paddle-imagenet-models-name.bj.bcebos.com/SE_ResNet34_vd_pretrained.tar)
- [SE_ResNet50_vd](https://paddle-imagenet-models-name.bj.bcebos.com/SE_ResNet50_vd_pretrained.tar)
@@ -90,7 +90,7 @@
- SE_ResNeXt_vd系列
- [SE_ResNeXt50_vd_32x4d](https://paddle-imagenet-models-name.bj.bcebos.com/SE_ResNeXt50_vd_32x4d_pretrained.tar)
- [SENet154_vd](https://paddle-imagenet-models-name.bj.bcebos.com/SENet154_vd_pretrained.tar)
- - Res2Net系列([论文地址](https://arxiv.org/abs/1904.01169))
+ - Res2Net系列[[9](#ref9)]([论文地址](https://arxiv.org/abs/1904.01169))
- [Res2Net50_26w_4s](https://paddle-imagenet-models-name.bj.bcebos.com/Res2Net50_26w_4s_pretrained.tar)
- [Res2Net50_vd_26w_4s](https://paddle-imagenet-models-name.bj.bcebos.com/Res2Net50_vd_26w_4s_pretrained.tar)
- [Res2Net50_14w_8s](https://paddle-imagenet-models-name.bj.bcebos.com/Res2Net50_14w_8s_pretrained.tar)
@@ -99,11 +99,11 @@
- Inception系列
- - GoogLeNet系列([论文地址](https://arxiv.org/pdf/1409.4842.pdf))
+ - GoogLeNet系列[[10](#ref10)]([论文地址](https://arxiv.org/pdf/1409.4842.pdf))
- [GoogLeNet](https://paddle-imagenet-models-name.bj.bcebos.com/GoogLeNet_pretrained.tar)
- - Inception系列([论文地址](https://arxiv.org/abs/1602.07261))
+ - Inception系列[[11](#ref11)]([论文地址](https://arxiv.org/abs/1602.07261))
- [InceptionV4](https://paddle-imagenet-models-name.bj.bcebos.com/InceptionV4_pretrained.tar)
- - Xception系列([论文地址](http://openaccess.thecvf.com/content_cvpr_2017/html/Chollet_Xception_Deep_Learning_CVPR_2017_paper.html))
+ - Xception系列[[12](#ref12)]([论文地址](http://openaccess.thecvf.com/content_cvpr_2017/html/Chollet_Xception_Deep_Learning_CVPR_2017_paper.html))
- [Xception41](https://paddle-imagenet-models-name.bj.bcebos.com/Xception41_pretrained.tar)
- [Xception41_deeplab](https://paddle-imagenet-models-name.bj.bcebos.com/Xception41_deeplab_pretrained.tar)
- [Xception65](https://paddle-imagenet-models-name.bj.bcebos.com/Xception65_pretrained.tar)
@@ -112,7 +112,7 @@
- HRNet系列
- - HRNet系列([论文地址](https://arxiv.org/abs/1908.07919))
+ - HRNet系列[[13](#ref13)]([论文地址](https://arxiv.org/abs/1908.07919))
- [HRNet_W18_C](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W18_C_pretrained.tar)
- [HRNet_W30_C](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W30_C_pretrained.tar)
- [HRNet_W32_C](https://paddle-imagenet-models-name.bj.bcebos.com/HRNet_W32_C_pretrained.tar)
@@ -123,13 +123,13 @@
- DPN与DenseNet系列
- - DPN系列([论文地址](https://arxiv.org/abs/1707.01629))
+ - DPN系列[[14](#ref14)]([论文地址](https://arxiv.org/abs/1707.01629))
- [DPN68](https://paddle-imagenet-models-name.bj.bcebos.com/DPN68_pretrained.tar)
- [DPN92](https://paddle-imagenet-models-name.bj.bcebos.com/DPN92_pretrained.tar)
- [DPN98](https://paddle-imagenet-models-name.bj.bcebos.com/DPN98_pretrained.tar)
- [DPN107](https://paddle-imagenet-models-name.bj.bcebos.com/DPN107_pretrained.tar)
- [DPN131](https://paddle-imagenet-models-name.bj.bcebos.com/DPN131_pretrained.tar)
- - DenseNet系列([论文地址](https://arxiv.org/abs/1608.06993))
+ - DenseNet系列[[15](#ref15)]([论文地址](https://arxiv.org/abs/1608.06993))
- [DenseNet121](https://paddle-imagenet-models-name.bj.bcebos.com/DenseNet121_pretrained.tar)
- [DenseNet161](https://paddle-imagenet-models-name.bj.bcebos.com/DenseNet161_pretrained.tar)
- [DenseNet169](https://paddle-imagenet-models-name.bj.bcebos.com/DenseNet169_pretrained.tar)
@@ -138,7 +138,7 @@
- EfficientNet与ResNeXt101_wsl系列
- - EfficientNet系列([论文地址](https://arxiv.org/abs/1905.11946))
+ - EfficientNet系列[[16](#ref16)]([论文地址](https://arxiv.org/abs/1905.11946))
- [EfficientNetB0_small](https://paddle-imagenet-models-name.bj.bcebos.com/EfficientNetB0_small_pretrained.tar)
- [EfficientNetB0](https://paddle-imagenet-models-name.bj.bcebos.com/EfficientNetB0_pretrained.tar)
- [EfficientNetB1](https://paddle-imagenet-models-name.bj.bcebos.com/EfficientNetB1_pretrained.tar)
@@ -148,7 +148,7 @@
- [EfficientNetB5](https://paddle-imagenet-models-name.bj.bcebos.com/EfficientNetB5_pretrained.tar)
- [EfficientNetB6](https://paddle-imagenet-models-name.bj.bcebos.com/EfficientNetB6_pretrained.tar)
- [EfficientNetB7](https://paddle-imagenet-models-name.bj.bcebos.com/EfficientNetB7_pretrained.tar)
- - ResNeXt101_wsl系列([论文地址](https://arxiv.org/abs/1805.00932))
+ - ResNeXt101_wsl系列[[17](#ref17)]([论文地址](https://arxiv.org/abs/1805.00932))
- [ResNeXt101_32x8d_wsl](https://paddle-imagenet-models-name.bj.bcebos.com/ResNeXt101_32x8d_wsl_pretrained.tar)
- [ResNeXt101_32x16d_wsl](https://paddle-imagenet-models-name.bj.bcebos.com/ResNeXt101_32x16d_wsl_pretrained.tar)
- [ResNeXt101_32x32d_wsl](https://paddle-imagenet-models-name.bj.bcebos.com/ResNeXt101_32x32d_wsl_pretrained.tar)
@@ -157,68 +157,69 @@
- 其他模型
- - AlexNet系列([论文地址](https://papers.nips.cc/paper/4824-imagenet-classification-with-deep-convolutional-neural-networks.pdf))
+ - AlexNet系列[[18](#ref18)]([论文地址](https://papers.nips.cc/paper/4824-imagenet-classification-with-deep-convolutional-neural-networks.pdf))
- [AlexNet](https://paddle-imagenet-models-name.bj.bcebos.com/AlexNet_pretrained.tar)
- - SqueezeNet系列([论文地址](https://arxiv.org/abs/1602.07360))
+ - SqueezeNet系列[[19](#ref19)]([论文地址](https://arxiv.org/abs/1602.07360))
- [SqueezeNet1_0](https://paddle-imagenet-models-name.bj.bcebos.com/SqueezeNet1_0_pretrained.tar)
- [SqueezeNet1_1](https://paddle-imagenet-models-name.bj.bcebos.com/SqueezeNet1_1_pretrained.tar)
- - VGG系列([论文地址](https://arxiv.org/abs/1409.1556))
+ - VGG系列[[20](#ref20)]([论文地址](https://arxiv.org/abs/1409.1556))
- [VGG11](https://paddle-imagenet-models-name.bj.bcebos.com/VGG11_pretrained.tar)
- [VGG13](https://paddle-imagenet-models-name.bj.bcebos.com/VGG13_pretrained.tar)
- [VGG16](https://paddle-imagenet-models-name.bj.bcebos.com/VGG16_pretrained.tar)
- [VGG19](https://paddle-imagenet-models-name.bj.bcebos.com/VGG19_pretrained.tar)
- - DarkNet系列([论文地址](https://arxiv.org/abs/1506.02640))
+ - DarkNet系列[[21](#ref21)]([论文地址](https://arxiv.org/abs/1506.02640))
- [DarkNet53](https://paddle-imagenet-models-name.bj.bcebos.com/DarkNet53_pretrained.tar)
- - ACNet系列([论文地址](https://arxiv.org/abs/1908.03930))
+ - ACNet系列[[22](#ref22)]([论文地址](https://arxiv.org/abs/1908.03930))
- [ResNet50_ACNet_deploy](https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_ACNet_deploy_pretrained.tar)
## 参考文献
-[1] He K, Zhang X, Ren S, et al. Deep residual learning for image recognition[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2016: 770-778.
-[2] He T, Zhang Z, Zhang H, et al. Bag of tricks for image classification with convolutional neural networks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2019: 558-567.
+[1] He K, Zhang X, Ren S, et al. Deep residual learning for image recognition[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2016: 770-778.
-[3] Howard A, Sandler M, Chu G, et al. Searching for mobilenetv3[C]//Proceedings of the IEEE International Conference on Computer Vision. 2019: 1314-1324.
+[2] He T, Zhang Z, Zhang H, et al. Bag of tricks for image classification with convolutional neural networks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2019: 558-567.
-[4] Sandler M, Howard A, Zhu M, et al. Mobilenetv2: Inverted residuals and linear bottlenecks[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2018: 4510-4520.
+[3] Howard A, Sandler M, Chu G, et al. Searching for mobilenetv3[C]//Proceedings of the IEEE International Conference on Computer Vision. 2019: 1314-1324.
-[5] Howard A G, Zhu M, Chen B, et al. Mobilenets: Efficient convolutional neural networks for mobile vision applications[J]. arXiv preprint arXiv:1704.04861, 2017.
+[4] Sandler M, Howard A, Zhu M, et al. Mobilenetv2: Inverted residuals and linear bottlenecks[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2018: 4510-4520.
-[6] Ma N, Zhang X, Zheng H T, et al. Shufflenet v2: Practical guidelines for efficient cnn architecture design[C]//Proceedings of the European Conference on Computer Vision (ECCV). 2018: 116-131.
+[5] Howard A G, Zhu M, Chen B, et al. Mobilenets: Efficient convolutional neural networks for mobile vision applications[J]. arXiv preprint arXiv:1704.04861, 2017.
-[7] Xie S, Girshick R, Dollár P, et al. Aggregated residual transformations for deep neural networks[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2017: 1492-1500.
+[6] Ma N, Zhang X, Zheng H T, et al. Shufflenet v2: Practical guidelines for efficient cnn architecture design[C]//Proceedings of the European Conference on Computer Vision (ECCV). 2018: 116-131.
+[7] Xie S, Girshick R, Dollár P, et al. Aggregated residual transformations for deep neural networks[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2017: 1492-1500.
-[8] Hu J, Shen L, Sun G. Squeeze-and-excitation networks[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2018: 7132-7141.
+[8] Hu J, Shen L, Sun G. Squeeze-and-excitation networks[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2018: 7132-7141.
-[9] Gao S, Cheng M M, Zhao K, et al. Res2net: A new multi-scale backbone architecture[J]. IEEE transactions on pattern analysis and machine intelligence, 2019.
-[10] Szegedy C, Liu W, Jia Y, et al. Going deeper with convolutions[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2015: 1-9.
+[9] Gao S, Cheng M M, Zhao K, et al. Res2net: A new multi-scale backbone architecture[J]. IEEE transactions on pattern analysis and machine intelligence, 2019.
+[10] Szegedy C, Liu W, Jia Y, et al. Going deeper with convolutions[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2015: 1-9.
-[11] Szegedy C, Ioffe S, Vanhoucke V, et al. Inception-v4, inception-resnet and the impact of residual connections on learning[C]//Thirty-first AAAI conference on artificial intelligence. 2017.
-[12] Chollet F. Xception: Deep learning with depthwise separable convolutions[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2017: 1251-1258.
+[11] Szegedy C, Ioffe S, Vanhoucke V, et al. Inception-v4, inception-resnet and the impact of residual connections on learning[C]//Thirty-first AAAI conference on artificial intelligence. 2017.
-[13] Wang J, Sun K, Cheng T, et al. Deep high-resolution representation learning for visual recognition[J]. arXiv preprint arXiv:1908.07919, 2019.
+[12] Chollet F. Xception: Deep learning with depthwise separable convolutions[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2017: 1251-1258.
-[14] Chen Y, Li J, Xiao H, et al. Dual path networks[C]//Advances in neural information processing systems. 2017: 4467-4475.
+[13] Wang J, Sun K, Cheng T, et al. Deep high-resolution representation learning for visual recognition[J]. arXiv preprint arXiv:1908.07919, 2019.
-[15] Huang G, Liu Z, Van Der Maaten L, et al. Densely connected convolutional networks[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2017: 4700-4708.
+[14] Chen Y, Li J, Xiao H, et al. Dual path networks[C]//Advances in neural information processing systems. 2017: 4467-4475.
+[15] Huang G, Liu Z, Van Der Maaten L, et al. Densely connected convolutional networks[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2017: 4700-4708.
-[16] Tan M, Le Q V. Efficientnet: Rethinking model scaling for convolutional neural networks[J]. arXiv preprint arXiv:1905.11946, 2019.
-[17] Mahajan D, Girshick R, Ramanathan V, et al. Exploring the limits of weakly supervised pretraining[C]//Proceedings of the European Conference on Computer Vision (ECCV). 2018: 181-196.
+[16] Tan M, Le Q V. Efficientnet: Rethinking model scaling for convolutional neural networks[J]. arXiv preprint arXiv:1905.11946, 2019.
-[18] Krizhevsky A, Sutskever I, Hinton G E. Imagenet classification with deep convolutional neural networks[C]//Advances in neural information processing systems. 2012: 1097-1105.
+[17] Mahajan D, Girshick R, Ramanathan V, et al. Exploring the limits of weakly supervised pretraining[C]//Proceedings of the European Conference on Computer Vision (ECCV). 2018: 181-196.
-[19] Iandola F N, Han S, Moskewicz M W, et al. SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and< 0.5 MB model size[J]. arXiv preprint arXiv:1602.07360, 2016.
+[18] Krizhevsky A, Sutskever I, Hinton G E. Imagenet classification with deep convolutional neural networks[C]//Advances in neural information processing systems. 2012: 1097-1105.
-[20] Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition[J]. arXiv preprint arXiv:1409.1556, 2014.
+[19] Iandola F N, Han S, Moskewicz M W, et al. SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and< 0.5 MB model size[J]. arXiv preprint arXiv:1602.07360, 2016.
-[21] Redmon J, Divvala S, Girshick R, et al. You only look once: Unified, real-time object detection[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2016: 779-788.
+[20] Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition[J]. arXiv preprint arXiv:1409.1556, 2014.
-[22] Ding X, Guo Y, Ding G, et al. Acnet: Strengthening the kernel skeletons for powerful cnn via asymmetric convolution blocks[C]//Proceedings of the IEEE International Conference on Computer Vision. 2019: 1911-1920.
+[21] Redmon J, Divvala S, Girshick R, et al. You only look once: Unified, real-time object detection[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2016: 779-788.
+
+[22] Ding X, Guo Y, Ding G, et al. Acnet: Strengthening the kernel skeletons for powerful cnn via asymmetric convolution blocks[C]//Proceedings of the IEEE International Conference on Computer Vision. 2019: 1911-1920.