[Doc] Add 2.5 release notes (#2030)

Author: aigcliu

README.md

@@ -20,19 +20,17 @@ English | [简体中文](README_CN.md)
 
 ## News <img src="./docs/images/seg_news_icon.png" width="40"/>
 <ul class="nobull">
-  <li>[2022-04-07] :fire: We add <a href="https://github.com/PaddlePaddle/PaddleSeg/tree/develop/configs/pp_liteseg">PP-LiteSeg</a>, a superior real-time semantic segmentation model, to develop branch. </li>
- <li>[2022-01-20] PaddleSeg v2.4 is released! More details in <a href="https://github.com/PaddlePaddle/PaddleSeg/releases">Release Notes</a>.</li>
- <ul>
-        <li>Upgrade the interactive annotation tool <a href="./EISeg">EISeg v0.4</a> with faster inference speed, supporting new areas in remote sensing and medical segmentation.</li>
-        <li>Publish the <a href="./contrib/PP-HumanSeg/paper.md">PP-HumanSeg</a> paper, and released Semantic Connectivity-aware Learning framework and a Large-Scale Teleconferencing Video Dataset.</li>
-        <li>Release <a href="./Matting">PP-HumanMatting</a> for extremely fine-grained human segmentation, with an Android tutorial.</li>
-        <li>Add 9 semantic segmentatioin models, including a new transformer model <a href="configs/segmenter">SegMenter</a>.</li>
-        <li>Provide a comprehensive <a href="docs/model_zoo_overview.md">performance table</a> for all segmentation models, which helps to choose a better model on the accuracy and the speed.</li>
+  <li>[2022-04-20] :fire: PaddleSeg v2.5 is released! More details in <a href="https://github.com/PaddlePaddle/PaddleSeg/releases">Release Notes</a>.</li>
+    <ul>
+        <li>Release <a href="./configs/pp_liteseg">PP-LiteSeg</a>, a real-time semantic segmentation model. It achieves SOTA trade-off between segmentation accuracy and inference speed. [<a href="https://arxiv.org/pdf/2204.02681.pdf">techical report</a>]</li>
+        <li>Release <a href="./Matting">PP-Matting</a>, a trimap-free image matting model for extremely fine-grained segmentation. It achieves SOTA performance on Composition-1k and Distinctions-646. [<a href="./Matting">techical report</a>]</li>
+        <li>Release <a href="./contrib/MedicalSeg">MedicalSeg</a>, a newly easy-to-use toolkit for 3D medical image segmentation. It supports the whole process including data preprocessing, model training, and model deployment, and provides the high-accuracy models on lung and spine segmentation.
+        <li>Upgrade the interactive annotation tool <a href="./EISeg">EISeg v0.5</a> with supporting new areas in chest X-Ray, MRI spine, and defect inspection.</li>
+        <li>Add 5 semantic segmentatioin models, including variants of PP-LiteSeg.</li>
     </ul>
-
-
+ <li>[2022-01-20] We release PaddleSeg v2.4 with EISeg v0.4, and <a href="./contrib/PP-HumanSeg">PP-HumanSeg</a> including open-sourced dataset <a href="./contrib/PP-HumanSeg/paper.md#pp-humanseg14k-a-large-scale-teleconferencing-video-dataset">PP-HumanSeg14K</a>. </li>
  <li>[2021-10-11] We released PaddleSeg v2.3 with the improved interactive segmentation tool EISeg v0.3, two matting algorithms, and segmentation model compression.</li>
- <li>[2021-09-20] We published an interactive segmentation <a href="https://arxiv.org/abs/2109.09406">paper</a> with SOTA performance on several well-known datasets.</li>
+
 </ul>
 
 
@@ -126,7 +124,11 @@ PaddleSeg is an end-to-end high-efficent development toolkit for image segmentat
             <li>MLA Transformer</li>
             <li>SegFormer</li>
             <li>SegMenter</li>
-            <li>PP-LiteSeg</li>
+            <li>ENet</li>
+            <li>CCNet</li>
+            <li>DDRNet</li>
+            <li>GloRe</li>
+            <li>PP-LiteSeg :star:</li>
       </td>
       <td>
         <b>Backbones</b><br>
@@ -159,6 +161,7 @@ PaddleSeg is an end-to-end high-efficent development toolkit for image segmentat
             <li>Edge Attention</li>
             <li>Relax Boundary</li>
             <li>Connectivity</li>
+            <li>MultiClassFocal</li>
           </ul>
         <b>Metrics</b><br>
           <ul>
@@ -337,11 +340,11 @@ PaddleSeg is released under the [Apache 2.0 license](LICENSE).
 * Thanks [yazheng0307](https://github.com/yazheng0307) (LIU Zheng) for contributing quick-start document.
 * Thanks [CuberrChen](https://github.com/CuberrChen) for contributing STDC(rethink BiSeNet), PointRend and DetailAggregateLoss.
 * Thanks [stuartchen1949](https://github.com/stuartchen1949) for contributing SegNet.
-* Thanks [justld](https://github.com/justld) (Lang Du) for contributing ESPNetV2, DMNet, ENCNet, HRNet_W48_Contrast, FastFCN, BiSeNetV1, SECrossEntropyLoss and PixelContrastCrossEntropyLoss.
+* Thanks [justld](https://github.com/justld) (Lang Du) for contributing DDRNet, CCNet, ESPNetV2, DMNet, ENCNet, HRNet_W48_Contrast, FastFCN, BiSeNetV1, SECrossEntropyLoss and PixelContrastCrossEntropyLoss.
 * Thanks [Herman-Hu-saber](https://github.com/Herman-Hu-saber) (Hu Huiming) for contributing ESPNetV2.
 * Thanks [zhangjin12138](https://github.com/zhangjin12138) for contributing RandomCenterCrop.
 * Thanks [simuler](https://github.com/simuler) for contributing ESPNetV1.
-* Thanks [ETTR123](https://github.com/ETTR123)(Zhang Kai) for contributing PFPNNet.
+* Thanks [ETTR123](https://github.com/ETTR123)(Zhang Kai) for contributing ENet, PFPNNet.
 
 ## Citation
 If you find our project useful in your research, please consider citing:

README_CN.md

@@ -19,14 +19,15 @@
 
 ## 最新动态 <img src="./docs/images/seg_news_icon.png" width="40"/>
 
-* [2022-01-20] PaddleSeg 2.4版本发布！详细发版信息请参考[Release Note](https://github.com/PaddlePaddle/PaddleSeg/releases)。
-  * 发布智能标注工具[EISeg v0.4](./EISeg)版，采用静态图预测加速交互过程，并支持遥感和医疗领域智能标注。
-  * 发表人像分割论文[PP-HumanSeg](./contrib/PP-HumanSeg/paper.md)，并开源连通性学习（SCL）方法和大规模视频会议数据集。
-  * 开源发丝级人像抠图Matting算法[PP-HumanMatting](./Matting)，并提供安卓端教程。
-  * 新增9个经典分割模型, 包括Transformer系列模型[SegMenter](configs/segmenter)，总模型数达到40个。
-  * 提供分割模型[性能对比图](docs/model_zoo_overview_cn.md)，全面展示分割模型性能，方便模型选型。
-* [2021-10-11] PaddleSeg 2.3版本发布交互式分割工具EISeg v0.3, 开源两种[Matting](./contrib/Matting)算法, 以及分割高阶功能[模型蒸馏](./slim/distill)和[模型量化](./slim/quant)方案。
-* [2021-09-20] PaddleSeg团队发表交互式分割论文[EdgeFlow](https://arxiv.org/abs/2109.09406)，已在多个数据集实现SOTA性能。
+* [2022-04-20] :fire: PaddleSeg 2.5版本发布！详细发版信息请参考[Release Note](https://github.com/PaddlePaddle/PaddleSeg/releases)。
+  * 发布超轻量级语义分割模型[PP-LiteSeg](./configs/pp_liteseg)以及[技术报告](https://arxiv.org/pdf/2204.02681.pdf)，实现精度和速度的最佳平衡。
+  * 发布高精度trimap-free抠图模型[PP-Matting](./Matting)以及技术报告，在Composition-1K和Distinctions-646上实现SOTA指标。
+  * 发布3D医疗影像开发套件[MedicalSeg](./contrib/MedicalSeg)，支持数据预处理、模型训练、模型部署等全流程开发，并提供肺部、椎骨数据上的高精度分割模型。
+  * 升级智能标注工具[EISeg v0.5](./EISeg)版，新增X-Ray胸腔标注、MRI椎骨标注、铝板瑕疵标注。
+  * 新增5个经典分割模型, 包括多个版本的PP-LiteSeg，总模型数达到45个。
+* [2022-01-20] PaddleSeg 2.4版本发布交互式分割工具EISeg v0.4，超轻量级人像分割方案[PP-HumanSeg](./contrib/PP-HumanSeg)，以及大规模视频会议数据集[PP-HumanSeg14K](./contrib/PP-HumanSeg/paper.md#pp-humanseg14k-a-large-scale-teleconferencing-video-dataset)。
+* [2021-10-11] PaddleSeg 2.3版本发布交互式分割工具EISeg v0.3，开源两种[Matting](./contrib/Matting)算法，以及分割高阶功能[模型蒸馏](./slim/distill)和[模型量化](./slim/quant)方案。
+
 
 ## 简介
 PaddleSeg是基于飞桨PaddlePaddle开发的端到端图像分割开发套件，涵盖了**高精度**和**轻量级**等不同方向的大量高质量分割模型。通过模块化的设计，提供了**配置化驱动**和**API调用**两种应用方式，帮助开发者更便捷地完成从训练到部署的全流程图像分割应用。
@@ -126,7 +127,11 @@ PaddleSeg是基于飞桨PaddlePaddle开发的端到端图像分割开发套件
             <li>MLA Transformer</li>
             <li>SegFormer</li>
             <li>SegMenter</li>
-            <li>PP-LiteSeg</li>
+            <li>ENet</li>
+            <li>CCNet</li>
+            <li>DDRNet</li>
+            <li>GloRe</li>
+            <li>PP-LiteSeg :star:</li>
       </td>
       <td>
         <b>骨干网络</b><br>
@@ -159,6 +164,7 @@ PaddleSeg是基于飞桨PaddlePaddle开发的端到端图像分割开发套件
             <li>Edge Attention</li>
             <li>Relax Boundary</li>
             <li>Connectivity</li>
+            <li>MultiClassFocal</li>
           </ul>
         <b>评估指标</b><br>
           <ul>
@@ -332,11 +338,11 @@ PaddleSeg是基于飞桨PaddlePaddle开发的端到端图像分割开发套件
 - 非常感谢[yazheng0307](https://github.com/yazheng0307) (刘正)贡献快速开始教程文档。
 - 非常感谢[CuberrChen](https://github.com/CuberrChen)贡献STDC (rethink BiSeNet) PointRend，和 Detail Aggregate损失函数。
 - 非常感谢[stuartchen1949](https://github.com/stuartchen1949)贡献 SegNet。
-- 非常感谢[justld](https://github.com/justld)(郎督)贡献 ESPNetV2, DMNet, ENCNet, HRNet_W48_Contrast, BiSeNetV1, FastFCN, SECrossEntropyLoss 和PixelContrastCrossEntropyLoss。
+- 非常感谢[justld](https://github.com/justld)(郎督)贡献 DDRNet, CCNet, ESPNetV2, DMNet, ENCNet, HRNet_W48_Contrast, BiSeNetV1, FastFCN, SECrossEntropyLoss 和PixelContrastCrossEntropyLoss。
 - 非常感谢[Herman-Hu-saber](https://github.com/Herman-Hu-saber)(胡慧明)参与贡献 ESPNetV2。
 - 非常感谢[zhangjin12138](https://github.com/zhangjin12138)贡献数据增强方法 RandomCenterCrop。
 - 非常感谢[simuler](https://github.com/simuler) 贡献 ESPNetV1。
-- 非常感谢[ETTR123](https://github.com/ETTR123)(张恺) 贡献 PFPNNet。
+- 非常感谢[ETTR123](https://github.com/ETTR123)(张恺) 贡献 ENet，PFPNNet。
 
 
 ## 学术引用 <img src="./docs/images/yinyong.png" width="30"/>

docs/model_zoo_overview.md

@@ -43,6 +43,11 @@ English | [简体中文](model_zoo_overview_cn.md)
 |[PFPNNet](../configs/pfpn)|-|✔|-|-|
 |[FastFCN](../configs/fastfcn)|✔|-|-|-|
 |[BiSeNetV1](../configs/bisenetv1)|-|-|-|-|
+|[ENet](../configs/enet)|-|-|-|-|
+|[CCNet](../configs/ccnet)|-|✔|-|-|
+|[DDRNet](../configs/ddrnet)|-|-|-|-|
+|[GloRe](../configs/glore)|✔|-|-|-|
+|[PP-LiteSeg](../configs/pp_liteseg)|-|-|-|-|
 
 ### Transformer series
 * [SETR](../configs/setr)

docs/model_zoo_overview_cn.md

@@ -43,6 +43,11 @@
 |[PFPNNet](../configs/pfpn)|-|✔|-|-|
 |[FastFCN](../configs/fastfcn)|✔|-|-|-|
 |[BiSeNetV1](../configs/bisenetv1)|-|-|-|-|
+|[ENet](../configs/enet)|-|-|-|-|
+|[CCNet](../configs/ccnet)|-|✔|-|-|
+|[DDRNet](../configs/ddrnet)|-|-|-|-|
+|[GloRe](../configs/glore)|✔|-|-|-|
+|[PP-LiteSeg](../configs/pp_liteseg)|-|-|-|-|
 
 ### Transformer系列
 * [SETR](../configs/setr)