Update mask2former-CN.md

mask2former-CN.md

@@ -1,6 +1,6 @@
 ---
 title: Universal Image Segmentation with Mask2Former and OneFormer
-thumbnail: /blog/assets/127_mask2former/thumbnail.png
+thumbnail: https://huggingface.co/blog/assets/127_mask2former/thumbnail.png
 authors:
 - user: nielsr
 - user: shivi
@@ -9,9 +9,7 @@ authors:
 
 # 通用图像分割：使用 Mask2Former 和 OneFormer
 
-**This guide introduces Mask2Former and OneFormer, 2 state-of-the-art neural networks for image segmentation. The models are now available in [`🤗 transformers`](https://huggingface.co/transformers), an open-source library that offers easy-to-use implementations of state-of-the-art models. Along the way, you'll learn about the difference between the various forms of image segmentation.**
-本文介绍两个领先的图像分割神经网络模型：Mask2Former和OneFormer。相关模型已经在[🤗 transformers](https://huggingface.co/transformers)提供。🤗 transformers 是一个开源库，提供了很多便捷的先进模型。在本文中，你也会学到各种图像分割任务的不同之处。
-
+**本文介绍两个领先的图像分割神经网络模型：Mask2Former和OneFormer。相关模型已经在[🤗 transformers](https://huggingface.co/transformers)提供。🤗 transformers 是一个开源库，提供了很多便捷的先进模型。在本文中，你也会学到各种图像分割任务的不同之处。**
 
 ## 图像分割