Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/1044197988/Centernet-Tensorflow2.0
Tensorflow2.0下运行目标检测网络Centernet(基于see--的keras-centernet)
https://github.com/1044197988/Centernet-Tensorflow2.0
centernet centernet-tensorflow2 keras-centernet object-detection tensorflow
Last synced: about 4 hours ago
JSON representation
Tensorflow2.0下运行目标检测网络Centernet(基于see--的keras-centernet)
- Host: GitHub
- URL: https://github.com/1044197988/Centernet-Tensorflow2.0
- Owner: 1044197988
- License: apache-2.0
- Created: 2019-09-18T07:20:51.000Z (about 5 years ago)
- Default Branch: master
- Last Pushed: 2023-03-24T22:34:36.000Z (over 1 year ago)
- Last Synced: 2024-08-02T19:01:45.789Z (3 months ago)
- Topics: centernet, centernet-tensorflow2, keras-centernet, object-detection, tensorflow
- Language: Python
- Size: 2.66 MB
- Stars: 37
- Watchers: 3
- Forks: 17
- Open Issues: 3
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
- Awesome-Tensorflow2 - 1044197988/Centernet-Tensorflow2.0
README
# Centernet-Tensorflow2.0
__Tensorflow2.0下运行目标检测网络Centernet(基于see--的keras-centernet)__
Centernet目标检测属于anchor-free系列的目标检测,相比于CornerNet做出了改进,使得检测速度和精度相比于one-stage和two-stage的框架都有不小的提高,尤其是与YOLOv3作比较,在相同速度的条件下,CenterNet的精度比YOLOv3提高了4个左右的点。![Centernet](https://github.com/1044197988/Centernet-Tensorflow2.0/blob/master/image/Centernet.png)
CenterNet是一种用于所有类型的对象检测相关任务的元算法。官方代码解决了2D检测,3D检测和人体姿势估计。对象不是常用的锚盒,而是表示为点。CenterNet还删除了以前单次探测器所需的许多超参数和概念:* 没有更多的锚箱
- 只有一个功能图表示所有比例
* 没有边界框匹配
- 没有非极大抑制![Centernet](https://github.com/1044197988/Centernet-Tensorflow2.0/blob/master/image/compare.png)
![Centernet](https://github.com/1044197988/Centernet-Tensorflow2.0/blob/master/image/image1.png)
![Centernet](https://github.com/1044197988/Centernet-Tensorflow2.0/blob/master/image/image2.png)权重下载:
---
点击网址或复制网址到浏览器链接即可下载,比较大,都在700MB以上.
对象检测:
https://github.com/see--/keras-centernet/releases/download/0.1.0/ctdet_coco_hg.hdf5
人体姿态估计:
https://github.com/see--/keras-centernet/releases/download/0.1.0/hpdet_coco_hg.hdf5## 参考
[see--/keras-centernet](https://github.com/see--/keras-centernet)
感谢see--的代码,修改很少,修改后可以在Tensorflow2.0下运行,在1版本包含keras下,也可以运行。
[kuku-sichuan/CenterNet](https://github.com/kuku-sichuan/CenterNet)## 其他框架下的Centernet
[xingyizhou-CenterNet-pytorch-官方](https://github.com/xingyizhou/CenterNet)
[Duankaiwen-CenterNet-pytorch](https://github.com/Duankaiwen/CenterNet)### 运行:
```Python
python TF2-CenterNet/ctdet_image.py #对象检测-图片
python TF2-CenterNet/hpdet_image.py #人体姿态估计-图片
python TF2-CenterNet/ctdet_video.py #对象检测-视频
python TF2-CenterNet/hpdet_video.py #人体姿态估计-视频
```
#在coco2017val数据集上执行测试-对象检测:
python TF2-CenterNet/ctdet_coco.py --data val2017 --annotations annotations
#在coco2017val数据集上执行测试-人体姿态估计:
python TF2-CenterNet/hpdet_coco.py --data val2017 --annotations annotations