update

4-Object_Detection/RPN/README.md

@@ -1,4 +1,4 @@
-## [RPN: RegionProposal Network](https://github.com/YunYang1994/ai-notebooks/blob/master/RPN.md)
+## [RPN: RegionProposal Network](https://github.com/YunYang1994/cv-notebooks/blob/master/ai_algorithm/RPN.md)
 --------------------
 
 This repository is implemented for paper ["Expecting the Unexpected: Training Detectors for Unusual Pedestrians with Adversarial Imposters(CVPR2017)"](https://arxiv.org/pdf/1703.06283), which makes some improvements on the basis of region proposal network in [Faster-RCNN](http://arxiv.org/abs/1506.01497). 

4-Object_Detection/YOLOV3/README.md

@@ -1,4 +1,4 @@
-## [TensorFlow2.x-YOLOv3](https://github.com/YunYang1994/ai-notebooks/blob/master/YOLOv3.md)
+## [TensorFlow2.x-YOLOv3](https://github.com/YunYang1994/cv-notebooks/blob/master/ai_algorithm/YOLOv3.md)
 --------------------
 A minimal tensorflow implementation of YOLOv3, with support for training, inference and evaluation.
 

5-Image_Segmentation/FCN/README.md

@@ -1,4 +1,4 @@
-## [FCN: Fully Convolutional Networks for Semantic Segmentation](https://github.com/YunYang1994/ai-notebooks/blob/master/FCN.md)
+## [FCN: Fully Convolutional Networks for Semantic Segmentation](https://github.com/YunYang1994/cv-notebooks/blob/master/ai_algorithm/FCN.md)
 --------------------
  [This paper](https://arxiv.org/abs/1411.4038) shows that convolutional networks by themselves, trained end-to-end, pixels-to-pixels, exceed the state-of-the-art in semantic segmentation. This repo adapts vgg16 networks into fully convolutional networks and transfer their learned representations by fine-tuning to the segmentation task on PASCAL VOC2012 dataset. 
 

5-Image_Segmentation/Unet/README.md

@@ -1,5 +1,5 @@
 
-## [U-Net: Convolutional Networks for Biomedical Image Segmentation.](https://github.com/YunYang1994/ai-notebooks/blob/master/Unet.md)
+## [U-Net: Convolutional Networks for Biomedical Image Segmentation.](https://github.com/YunYang1994/cv-notebooks/blob/master/ai_algorithm/Unet.md)
 --------------------
  [this paper](https://arxiv.org/abs/1505.04597) presents a network and training strategy that relies on the strong use of data augmentation to use the available annotated samples more efficiently. The architecture consists of a contracting path to capture context and a symmetric expanding path that enables precise localization.