Skip to content

P
PaddleDetection

PaddlePaddle / PaddleDetection
大约 1 年 前同步成功

通知 695
Star 11112
Fork 2696
P
PaddleDetection
未验证 提交 b7748ee2 编写于 作者: K Kaipeng Deng 提交者： GitHub
浏览文件
操作

[Cherry pick] update readme (#4410) 

* chang image
上级 6e65f11d
隐藏空白更改
内联 并排
Showing with 108 addition and 31 deletion
README_cn.md
浏览文件 @ b7748ee2
......@@ -4,7 +4,7 @@
# 产品动态
- 2021.11.03: 发布[release/2.3版本](https://github.com/PaddlePaddle/PaddleDetection/tree/release/2.3)，发布**轻量级检测特色模型**⚡[PP-PicoDet](configs/picodet)，发布**轻量级关键点特色模型**⚡[PP-TinyPose](configs/keypoint/tiny_pose)。新增[Swin Transformer](configs/faster_rcnn)，[TOOD](configs/tood)，[GFL](configs/gfl)目标检测模型。发布[Sniper](configs/sniper)小目标检测优化模型，发布针对EdgeBoard优化[PP-YOLO-EB](configs/ppyolo)模型。新增轻量化关键点模型[Lite HRNet](configs/keypoint)关键点模型并支持Paddle Lite部署。
- 2021.08.10: 发布[release/2.2版本]((https://github.com/PaddlePaddle/PaddleDetection/tree/release/2.2))，发布Transformer检测系列模型，包括[DETR](configs/detr), [Deformable DETR](configs/deformable_detr), [Sparse RCNN](configs/sparse_rcnn)。新增Dark HRNet关键点模型和MPII数据集[关键点模型](configs/keypoint)，新增[人头](configs/mot/headtracking21)、[车辆](configs/mot/vehicle)跟踪垂类模型。
- 2021.08.10: 发布[release/2.2版本](https://github.com/PaddlePaddle/PaddleDetection/tree/release/2.2)，发布Transformer检测系列模型，包括[DETR](configs/detr), [Deformable DETR](configs/deformable_detr), [Sparse RCNN](configs/sparse_rcnn)。新增Dark HRNet关键点模型和MPII数据集[关键点模型](configs/keypoint)，新增[人头](configs/mot/headtracking21)、[车辆](configs/mot/vehicle)跟踪垂类模型。
- 2021.05.20: 发布[release/2.1版本](https://github.com/PaddlePaddle/PaddleDetection/tree/release/2.1)，新增[关键点检测](configs/keypoint)，模型包括HigherHRNet，HRNet。新增[多目标跟踪](configs/mot)能力，模型包括DeepSORT，JDE，FairMOT。发布PPYOLO系列模型压缩模型，新增[ONNX模型导出教程](deploy/EXPORT_ONNX_MODEL.md)。
# 简介
......@@ -13,13 +13,8 @@ PaddleDetection为基于飞桨PaddlePaddle的端到端目标检测套件，提
### PaddleDetection提供了目标检测、实例分割、多目标跟踪、关键点检测等多种能力
<div width="900" align="center">
<img src="docs/images/det.jpg" width="400" height="300" title="目标检测"/>
<img src="docs/images/ins.jpg" width="400" height="300" title="实例分割"/>
</div>
<div width="900" align="center">
<img src="docs/images/mot.gif" width="400" height="300" title="多目标跟踪"/>
<img src="docs/images/pose.gif" width="400" height="300" title="关键点检测"/>
<div width="1000" align="center">
<img src="docs/images/ppdet.gif"/>
</div>
......@@ -33,7 +28,7 @@ PaddleDetection为基于飞桨PaddlePaddle的端到端目标检测套件，提
### 套件结构概览
<table>
<table align="center">
<tbody>
<tr align="center" valign="bottom">
<td>
......@@ -66,7 +61,7 @@ PaddleDetection为基于飞桨PaddlePaddle的端到端目标检测套件，提
<li>RetinaNet</li>
<li>YOLOv3</li>
<li>YOLOv4</li>
<li>PP-YOLO</li>
<li>PP-YOLOv1/v2/Tiny</li>
<li>SSD</li>
</ul>
</ul>
......@@ -75,6 +70,7 @@ PaddleDetection为基于飞桨PaddlePaddle的端到端目标检测套件，提
<li>CornerNet-Squeeze</li>
<li>FCOS</li>
<li>TTFNet</li>
<li>PicoDet</li>
</ul>
</ul>
<ul>
......@@ -92,6 +88,29 @@ PaddleDetection为基于飞桨PaddlePaddle的端到端目标检测套件，提
<li>BlazeFace-NAS</li>
</ul>
</ul>
<ul>
<li><b>Transformer</b></li>
<ul>
<li>DETR/Deformable DETR</li>
<li>Sparse RCNN</li>
<li>Swin Transformer</li>
</ul>
</ul>
<ul>
<li><b>Multi-Object-Tracking</b></li>
<ul>
<li>JDE</li>
<li>FairMOT</li>
<li>DeepSort</li>
</ul>
</ul>
<ul>
<li><b>KeyPoint-Detection</b></li>
<ul>
<li>HRNet</li>
<li>HigherHRNet</li>
</ul>
</ul>
</td>
<td>
<ul>
......@@ -109,6 +128,7 @@ PaddleDetection为基于飞桨PaddlePaddle的端到端目标检测套件，提
<li>MobileNetv1/v3</li>
<li>GhostNet</li>
<li>Efficientnet</li>
<li>BlazeNet</li>
</ul>
</td>
<td>
......@@ -120,6 +140,11 @@ PaddleDetection为基于飞桨PaddlePaddle的端到端目标检测套件，提
<li>Non-local</li>
</ul>
</ul>
<ul><li><b>KeyPoint</b></li>
<ul>
<li>DarkPose</li>
</ul>
</ul>
<ul><li><b>FPN</b></li>
<ul>
<li>BiFPN</li>
......@@ -151,15 +176,18 @@ PaddleDetection为基于飞桨PaddlePaddle的端到端目标检测套件，提
<td>
<ul>
<li>Resize</li>
<li>Lighting</li>
<li>Flipping</li>
<li>Expand</li>
<li>Crop</li>
<li>Color Distort</li>
<li>Random Erasing</li>
<li>Mixup </li>
<li>Mosaic</li>
<li>Cutmix </li>
<li>Grid Mask</li>
<li>Auto Augment</li>
<li>Random Perspective</li>
</ul>
</td>
</tr>
......@@ -186,6 +214,16 @@ PaddleDetection为基于飞桨PaddlePaddle的端到端目标检测套件，提
- `PP-YOLO v2`是对`PP-YOLO`模型的进一步优化，在COCO数据集精度49.5%，Tesla V100预测速度68.9FPS
- 图中模型均可在[模型库](#模型库)中获取
各移动端模型在COCO数据集上精度mAP和高通骁龙865处理器上预测速度(FPS)对比图。
<div align="center">
<img src="docs/images/mobile_fps_map.png" width=600/>
</div>
**说明：**
- 测试数据均使用高通骁龙865(4\*A77 + 4\*A55)处理器batch size为1, 开启4线程测试，测试使用NCNN预测库，测试脚本见[MobileDetBenchmark](https://github.com/JiweiMaster/MobileDetBenchmark)
- [PP-PicoDet](configs/picodet)及[PP-YOLO-Tiny](configs/ppyolo)为PaddleDetection自研模型，其余模型PaddleDetection暂未提供
## 文档教程
### 入门教程
......
README_en.md
浏览文件 @ b7748ee2
......@@ -3,8 +3,8 @@ English | [简体中文](README_cn.md)
# Product news
- 2021.11.03: Release [release/2.3]((https://github.com/PaddlePaddle/Paddleetection/tree/release/2.3) version. Release mobile object detection model ⚡[PP-PicoDet](configs/picodet), mobile keypoint detection model ⚡[PP-TinyPose](configs/keypoint/tiny_pose). Release object detection models, including [Swin-Transformer](configs/faster_rcnn), [TOOD](configs/tood), [GFL](configs/gfl), release [Sniper](configs/sniper) tiny object detection models and optimized [PP-YOLO-EB](configs/ppyolo) model for EdgeBoard. Release mobile keypoint detection model [Lite HRNet](configs/keypoint).
- 2021.08.10: Release [release/2.2]((https://github.com/PaddlePaddle/Paddleetection/tree/release/2.2) version. Release Transformer object detection models, including [DETR](configs/detr), [Deformable DETR](configs/deformable_detr), [Sparse RCNN](configs/sparse_rcnn). Release [keypoint detection](configs/keypoint) models, including DarkHRNet and model trained on MPII dataset. Release [head-tracking](configs/mot/headtracking21) and [vehicle-tracking](configs/mot/vehicle) multi-object tracking models.
- 2021.11.03: Release [release/2.3](https://github.com/PaddlePaddle/Paddleetection/tree/release/2.3) version. Release mobile object detection model ⚡[PP-PicoDet](configs/picodet), mobile keypoint detection model ⚡[PP-TinyPose](configs/keypoint/tiny_pose). Release object detection models, including [Swin-Transformer](configs/faster_rcnn), [TOOD](configs/tood), [GFL](configs/gfl), release [Sniper](configs/sniper) tiny object detection models and optimized [PP-YOLO-EB](configs/ppyolo) model for EdgeBoard. Release mobile keypoint detection model [Lite HRNet](configs/keypoint).
- 2021.08.10: Release [release/2.2](https://github.com/PaddlePaddle/Paddleetection/tree/release/2.2) version. Release Transformer object detection models, including [DETR](configs/detr), [Deformable DETR](configs/deformable_detr), [Sparse RCNN](configs/sparse_rcnn). Release [keypoint detection](configs/keypoint) models, including DarkHRNet and model trained on MPII dataset. Release [head-tracking](configs/mot/headtracking21) and [vehicle-tracking](configs/mot/vehicle) multi-object tracking models.
- 2021.05.20: Release [release/2.1]((https://github.com/PaddlePaddle/Paddleetection/tree/release/2.1) version. Release [Keypoint Detection](configs/keypoint), including HigherHRNet and HRNet, [Multi-Object Tracking](configs/mot), including DeepSORT，JDE and FairMOT. Release model compression for PPYOLO series models.Update documents such as [EXPORT ONNX MODEL](deploy/EXPORT_ONNX_MODEL.md).
......@@ -14,13 +14,8 @@ PaddleDetection is an end-to-end object detection development kit based on Paddl
### PaddleDetection provides image processing capabilities such as object detection, instance segmentation, multi-object tracking, keypoint detection and etc.
<div width="900" align="center">
<img src="docs/images/det.jpg" width="400" height="300" title="目标检测"/>
<img src="docs/images/ins.jpg" width="400" height="300" title="实例分割"/>
</div>
<div width="900" align="center">
<img src="docs/images/mot.gif" width="400" height="300" title="多目标跟踪"/>
<img src="docs/images/pose.gif" width="400" height="300" title="关键点检测"/>
<div width="1000" align="center">
<img src="docs/images/ppdet.gif"/>
</div>
......@@ -40,7 +35,7 @@ Based on the high performance core of PaddlePaddle, advantages of training speed
#### Overview of Kit Structures
<table>
<table align="center">
<tbody>
<tr align="center" valign="bottom">
<td>
......@@ -65,7 +60,7 @@ Based on the high performance core of PaddlePaddle, advantages of training speed
<li>Cascade-RCNN</li>
<li>Libra RCNN</li>
<li>Hybrid Task RCNN</li>
<li>PSS-Det RCNN</li>
<li>PSS-Det</li>
</ul>
</ul>
<ul><li><b>One-Stage Detection</b></li>
......@@ -73,7 +68,7 @@ Based on the high performance core of PaddlePaddle, advantages of training speed
<li>RetinaNet</li>
<li>YOLOv3</li>
<li>YOLOv4</li>
<li>PP-YOLO</li>
<li>PP-YOLOv1/v2/Tiny</li>
<li>SSD</li>
</ul>
</ul>
......@@ -82,6 +77,7 @@ Based on the high performance core of PaddlePaddle, advantages of training speed
<li>CornerNet-Squeeze</li>
<li>FCOS</li>
<li>TTFNet</li>
<li>PicoDet</li>
</ul>
</ul>
<ul>
......@@ -92,13 +88,36 @@ Based on the high performance core of PaddlePaddle, advantages of training speed
</ul>
</ul>
<ul>
<li><b>Face-Detection</b></li>
<li><b>Face-Detction</b></li>
<ul>
<li>FaceBoxes</li>
<li>BlazeFace</li>
<li>BlazeFace-NAS</li>
</ul>
</ul>
<ul>
<li><b>Transformer</b></li>
<ul>
<li>DETR/Deformable DETR</li>
<li>Sparse RCNN</li>
<li>Swin Transformer</li>
</ul>
</ul>
<ul>
<li><b>Multi-Object-Tracking</b></li>
<ul>
<li>JDE</li>
<li>FairMOT</li>
<li>DeepSort</li>
</ul>
</ul>
<ul>
<li><b>KeyPoint-Detection</b></li>
<ul>
<li>HRNet</li>
<li>HigherHRNet</li>
</ul>
</ul>
</td>
<td>
<ul>
......@@ -116,6 +135,7 @@ Based on the high performance core of PaddlePaddle, advantages of training speed
<li>MobileNetv1/v3</li>
<li>GhostNet</li>
<li>Efficientnet</li>
<li>BlazeNet</li>
</ul>
</td>
<td>
......@@ -127,6 +147,11 @@ Based on the high performance core of PaddlePaddle, advantages of training speed
<li>Non-local</li>
</ul>
</ul>
<ul><li><b>KeyPoint</b></li>
<ul>
<li>DarkPose</li>
</ul>
</ul>
<ul><li><b>FPN</b></li>
<ul>
<li>BiFPN</li>
......@@ -158,15 +183,18 @@ Based on the high performance core of PaddlePaddle, advantages of training speed
<td>
<ul>
<li>Resize</li>
<li>Lighting</li>
<li>Flipping</li>
<li>Expand</li>
<li>Crop</li>
<li>Color Distort</li>
<li>Random Erasing</li>
<li>Mixup </li>
<li>Mosaic</li>
<li>Cutmix </li>
<li>Grid Mask</li>
<li>Auto Augment</li>
<li>Random Perspective</li>
</ul>
</td>
</tr>
......@@ -178,24 +206,35 @@ Based on the high performance core of PaddlePaddle, advantages of training speed
</table>
#### Overview of Model Performance
The relationship between COCO mAP and FPS on Tesla V100 of representative models of each architectures and backbones.
The relationship between COCO mAP and FPS on Tesla V100 of representative models of each server side architectures and backbones.
<div align="center">
<img src="docs/images/fps_map.png" />
</div>
</div>
**NOTE:**
**NOTE:**
- `CBResNet stands` for `Cascade-Faster-RCNN-CBResNet200vd-FPN`, which has highest mAP on COCO as 53.3%
- `CBResNet stands` for `Cascade-Faster-RCNN-CBResNet200vd-FPN`, which has highest mAP on COCO as 53.3%
- `Cascade-Faster-RCNN` stands for `Cascade-Faster-RCNN-ResNet50vd-DCN`, which has been optimized to 20 FPS inference speed when COCO mAP as 47.8% in PaddleDetection models
- `Cascade-Faster-RCNN` stands for `Cascade-Faster-RCNN-ResNet50vd-DCN`, which has been optimized to 20 FPS inference speed when COCO mAP as 47.8% in PaddleDetection models
- `PP-YOLO` achieves mAP of 45.9% on COCO and 72.9FPS on Tesla V100. Both precision and speed surpass [YOLOv4](https://arxiv.org/abs/2004.10934)
- `PP-YOLO` achieves mAP of 45.9% on COCO and 72.9FPS on Tesla V100. Both precision and speed surpass [YOLOv4](https://arxiv.org/abs/2004.10934)
- `PP-YOLO v2` is optimized version of `PP-YOLO` which has mAP of 49.5% and 68.9FPS on Tesla V100
- `PP-YOLO v2` is optimized version of `PP-YOLO` which has mAP of 49.5% and 68.9FPS on Tesla V100
- All these models can be get in [Model Zoo](#Model-Zoo)
- All these models can be get in [Model Zoo](#ModelZoo)
The relationship between COCO mAP and FPS on Qualcomm Snapdragon 865 of representative mobile side models.
<div align="center">
<img src="docs/images/mobile_fps_map.png" width=600 />
</div>
**NOTE:**
- All data tested on Qualcomm Snapdragon 865(4\*A77 + 4\*A55) processor with batch size of 1 and CPU threads of 4, and use NCNN library in testing, benchmark scripts is publiced at [MobileDetBenchmark](https://github.com/JiweiMaster/MobileDetBenchmark)
- [PP-PicoDet](configs/picodet) and [PP-YOLO-Tiny](configs/ppyolo) are developed and released by PaddleDetection, other models are not provided in PaddleDetection.
## Tutorials
......
Markdown is supported
0% 或 .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑！
想要评论请 注册 或