Skip to content

Commit

Permalink
release_rdm
Browse files Browse the repository at this point in the history
  • Loading branch information
@shiyutang
shiyutang committed Oct 18, 2023
1 parent 3c3a12a commit af455f7
Show file tree
Hide file tree
Showing 2 changed files with 11 additions and 10 deletions.
11 changes: 6 additions & 5 deletions README_CN.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -22,11 +22,12 @@

## <img src="./docs/images/seg_news_icon.png" width="20"/> 最新动态

* [2023-04-11] :fire: PaddleSeg 2.8版本发布！详细发版信息请参考[Release Note](https://github.com/PaddlePaddle/PaddleSeg/releases)。
* 开源飞桨版本视觉大模型[Segment Anything Model](./contrib/SegmentAnything)，提供演示Demo，支持全图分割的功能、使用提示输入分割的功能。
* 发布超轻量级语义分割SOTA模型[PP-MobileSeg](./configs/pp_mobileseg)：在ADE20K数据集上，PP-MobileSeg对比此前最优模型，精度提升1.5%、速度提升42.3%、参数量减少34.9%。
* 发布工业质检全流程解决方案[QualityInspector v0.5](./contrib/QualityInspector)：提供统一可配置的算法Pipeline，集成检测和分割模型库，支持3种无监督异常检测算法，支持工业级指标评测、分析、调优等功能。
* 发布通用的全景分割解决方案[PanopticSeg v0.5](./contrib/PanopticSeg)：提供全景分割的全流程开发功能，集成2个前沿高精模型，具备灵活的二次开发能力。
* [2023-10-29] :fire: PaddleSeg 2.9版本发布！详细发版信息请参考[Release Note](https://github.com/PaddlePaddle/PaddleSeg/releases)。
* 增加对多标签分割[Multi-label segmentation](./configs/multilabelseg)，提供数据转换代码及结果可视化，实现对一系列语义分割模型的多标签分割支持。
* 发布轻量视觉大模型[MobileSAM](./configs/SegmentAnything)，实现更快速的SAM推理。
* 支持量化蒸馏训练压缩功能[Quant Aware Distillation Training Compression](./deploy/slim/act)，对PP-LiteSeg、PP-MobileSeg、OCRNet、SegFormer-B0增加量化训练压缩功能，提升推理速度。

* [2022-04-11] PaddleSeg 2.8版本发布视觉大模型[Segment Anything Model](./contrib/SegmentAnything)、轻量级语义分割SOTA模型[PP-MobileSeg](./configs/pp_mobileseg)、工业质检全流程解决方案[QualityInspector v0.5](./contrib/QualityInspector)、通用的全景分割解决方案[PanopticSeg v0.5](./contrib/PanopticSeg)。
* [2022-11-30] PaddleSeg 2.7版本发布实时人像抠图模型[PP-MattingV2](./Matting/)、3D医疗影像分割方案[MedicalSegV2](./contrib/MedicalSeg/)、轻量级语义分割模型[RTFormer](./configs/rtformer/)。
* [2022-07-20] PaddleSeg 2.6版本发布实时人像分割SOTA方案[PP-HumanSegV2](./contrib/PP-HumanSeg)、高性能智能标注工具[EISeg v1.0](./EISeg)正式版、ImageNet分割伪标签数据预训练方法PSSL，开源PP-MattingV1代码和预训练模型。
* [2022-04-20] PaddleSeg 2.5版本发布超轻量级语义分割模型[PP-LiteSeg](./configs/pp_liteseg)，高精度抠图模型PP-MattingV1，3D医疗影像开发套件MedicalSegV1，交互式分割工具EISeg v0.5。
Expand Down
10 changes: 5 additions & 5 deletions README_EN.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -23,13 +23,13 @@ English | [简体中文](README_CN.md)

## <img src="./docs/images/seg_news_icon.png" width="20"/> News
<ul class="nobull">
<li>[2023-04-11] :fire: PaddleSeg v2.8 is released! Check more details in <a href="https://github.com/PaddlePaddle/PaddleSeg/releases">Release Notes</a>.</li>
<li>[2023-10-29] :fire: PaddleSeg v2.9 is released! Check more details in <a href="https://github.com/PaddlePaddle/PaddleSeg/releases">Release Notes</a>.</li>
<ul>
<li>Release <a href="./contrib/SegmentAnything">Segment Anything Model</a> based on PaddlePaddle. Demos are provided to demonstrate the function of automatic full-image segmentation and specified object segmentation with prompt input.</li>
<li>Release <a href="./configs/pp_mobileseg">PP-MobileSeg</a>, a lightweight semantic segmentation model for mobile devices. Comparing PP-MobileSeg with other models on the ADE20K dataset, the segmentation accuracy is improved by 1.5%, the inference speed is accelerated by 42.3%, and the number of parameters is decreased by 34.9%. </li>
<li>Release <a href="./contrib/QualityInspector">QualityInspector v0.5</a>, a full-process solution for industrial quality inspection. It provides a unified and configurable pipeline for single-task and multi-task models, integrates detection and segmentation model libraries, and supports three unsupervised quality inspection methods. </li>
<li>Release <a href="./contrib/PanopticSeg">PanopticSeg v0.5</a>, a universal panoptic segmentation solution. It provides the full-process capabilities of panoptic segmentation, integrates two models, and has flexible secondary development capabilities. </li>
<li>Support <a href="./configs/multilabelseg">Multi-label segmentation</a>, it procides multi-label segmentation support on a serie of semantic segmetation models.</li>
<li>Release <a href="./configs/SegmentAnything">Mobile SAM</a>, faster version of Segment Anything Model. </li>
<li>Support <a href="./deploy/slim/act">Quant Aware Distillation Training Compression</a> for PP-LiteSeg, PP-MobileSeg, OCRNet, and SegFormer-B0 to improve model inference speed. </li>
</ul>
<li>[2022-04-11] PaddleSeg v2.8 released <a href="./contrib/SegmentAnything">Segment Anything Model</a>, an original light-weight semantic segmentation model on mobile devices <a href="./configs/pp_mobileseg">PP-MobileSeg</a>, <a href="./contrib/QualityInspector">QualityInspector v0.5</a>, a full-process solution for industrial quality inspection, and <a href="./contrib/PanopticSeg">PanopticSeg v0.5</a>, a universal panoptic segmentation solution.
<li>[2022-11-30] PaddleSeg v2.7 released a real-time human matting model <a href="./Matting/">PP-MattingV2</a>, a 3D medical image segmentation solution <a href="./contrib/MedicalSeg/">MedicalSegV2</a>, and a real-time semantic segmentation model <a href="./configs/rtformer/">RTFormer</a>.
<li>[2022-07-20] PaddleSeg v2.6 released a real-time human segmentation SOTA solution <a href="./contrib/PP-HumanSeg">PP-HumanSegV2</a>, a stable-version semi-automatic segmentation annotation tool <a href="./EISeg">EISeg v1.0</a>, a pseudo label pre-training method PSSL, and the source code of PP-MattingV1. </li>
<li>[2022-04-20] PaddleSeg v2.5 released a real-time semantic segmentation model <a href="./configs/pp_liteseg">PP-LiteSeg</a>, a trimap-free image matting model PP-MattingV1, and an easy-to-use solution for 3D medical image segmentation MedicalSegV1.</li>
Expand Down

0 comments on commit af455f7

Please sign in to comment.