Skip to content

Commit

Permalink
更新模型加载说明
Browse files Browse the repository at this point in the history
  • Loading branch information
mymagicpower committed Nov 13, 2024
1 parent ea4d048 commit 458314a
Show file tree
Hide file tree
Showing 92 changed files with 913 additions and 104 deletions.
71 changes: 50 additions & 21 deletions 1_image_sdks/classification/animal_sdk/README.md
Original file line number Diff line number Diff line change
@@ -1,24 +1,31 @@
### 官网:
[官网链接](https://www.aias.top/)

### Download the model and place it in the /models directory
- Link 1: https://github.com/mymagicpower/AIAS/releases/download/apps/animals.zip
- Link 2: https://github.com/mymagicpower/AIAS/releases/download/apps/mobilenet_animals.zip
-
### Animal classification recognition SDK
Animal recognition SDK that supports the classification of 7978 types of animals.
### 下载模型
- 链接: https://pan.baidu.com/s/1LLwbo3Wvu96c1lID4drgoQ?pwd=41mj

### ### SDK features
- Supports the classification recognition of 7978 types of animals and provides confidence levels.
- Provides two available model examples:
1). Example of the large model (resnet50): AnimalsClassificationExample
2). Example of the small model (mobilenet_v2): LightAnimalsClassExample
### 模型使用方法:
- 1. 用模型的名字搜索代码,找到模型的加载位置
- 2. 然后更新模型路径(代码里默认加载路径是:项目/models 文件夹)
- 3. 具体模型加载方法
- http://aias.top/AIAS/guides/load_model.html

[Animal classification](https://aias-home.oss-cn-beijing.aliyuncs.com/AIAS/animal_sdk/animals.txt)
### 动物分类识别SDK
动物识别sdk,支持7978种动物的分类识别。

### Running examples
- Test image
![tiger](https://aias-home.oss-cn-beijing.aliyuncs.com/AIAS/animal_sdk/tiger.jpeg)
### SDK功能
- 支持7978种动物的分类识别,并给出置信度。
- 提供两个可用模型例子
1). 大模型(resnet50)例子:AnimalsClassificationExample
2). 小模型(mobilenet_v2)例子:LightAnimalsClassExample

After successful execution, the command line should display the following information:
[动物分类](https://aias-home.oss-cn-beijing.aliyuncs.com/AIAS/animal_sdk/animals.txt)

### 运行例子
- 测试图片
![tiger](https://aias-home.oss-cn-beijing.aliyuncs.com/AIAS/animal_sdk/tiger.jpeg)

运行成功后,命令行应该看到下面的信息:
```text
西伯利亚虎 : 1.0
[INFO ] - [
Expand All @@ -31,9 +38,31 @@ After successful execution, the command line should display the following inform
```


### Open source algorithms
#### 1. Open source algorithm used in the SDK
### 开源算法
#### 1. sdk使用的开源算法
- [PaddleClas](https://github.com/PaddlePaddle/PaddleClas)
#### 2. How to export the model?
- [export_model](https://github.com/PaddlePaddle/PaddleClas/blob/release%2F2.2/tools/export_model.py)
- [how_to_create_paddlepaddle_model](http://docs.djl.ai/docs/paddlepaddle/how_to_create_paddlepaddle_model_zh.html)
#### 2. 模型如何导出 ?
- [export_model](https://github.com/PaddlePaddle/PaddleClas/blob/release%2F2.2/tools/export_model.py)
- [how_to_create_paddlepaddle_model](http://docs.djl.ai/docs/paddlepaddle/how_to_create_paddlepaddle_model_zh.html)



### 其它帮助信息
https://aias.top/guides.html

### Git地址:
[Github链接](https://github.com/mymagicpower/AIAS)
[Gitee链接](https://gitee.com/mymagicpower/AIAS)



#### 帮助文档:
- https://aias.top/guides.html
- 1.性能优化常见问题:
- https://aias.top/AIAS/guides/performance.html
- 2.引擎配置(包括CPU,GPU在线自动加载,及本地配置):
- https://aias.top/AIAS/guides/engine_config.html
- 3.模型加载方式(在线自动加载,及本地配置):
- https://aias.top/AIAS/guides/load_model.html
- 4.Windows环境常见问题:
- https://aias.top/AIAS/guides/windows.html
8 changes: 7 additions & 1 deletion 1_image_sdks/classification/animal_sdk/README_cn.md
Original file line number Diff line number Diff line change
@@ -1,9 +1,15 @@
### 官网:
[官网链接](https://www.aias.top/)

### 下载模型,放置于models目录
### 下载模型
- 链接: https://pan.baidu.com/s/1LLwbo3Wvu96c1lID4drgoQ?pwd=41mj

### 模型使用方法:
- 1. 用模型的名字搜索代码,找到模型的加载位置
- 2. 然后更新模型路径(代码里默认加载路径是:项目/models 文件夹)
- 3. 具体模型加载方法
- http://aias.top/AIAS/guides/load_model.html

### 动物分类识别SDK
动物识别sdk,支持7978种动物的分类识别。

Expand Down
62 changes: 44 additions & 18 deletions 1_image_sdks/classification/dish_sdk/README.md
Original file line number Diff line number Diff line change
@@ -1,25 +1,31 @@
### 官网:
[官网链接](https://www.aias.top/)

###Download models and place them in the /models directory
### 下载模型
- 链接: https://pan.baidu.com/s/1No67j8xhXlDfx676P14yQw?pwd=gtdq

- Link 1: https://github.com/mymagicpower/AIAS/releases/download/apps/dishes.zip
- Link 2: https://github.com/mymagicpower/AIAS/releases/download/apps/mobilenet_dishes.zip
### 模型使用方法:
- 1. 用模型的名字搜索代码,找到模型的加载位置
- 2. 然后更新模型路径(代码里默认加载路径是:项目/models 文件夹)
- 3. 具体模型加载方法
- http://aias.top/AIAS/guides/load_model.html

### Dish Classification Recognition SDK
The dish recognition SDK supports the classification recognition of 8416 kinds of dishes.
### 菜品分类识别SDK
菜品识别sdk,支持8416种菜品的分类识别。

### SDK Functions
-Supports the classification recognition of 8416 kinds of dishes and provides confidence levels.
-Provides two available model examples
1. Example of the large model (resnet50): DishesClassificationExample
2. Example of the small model (mobilenet_v2): LightDishesClassExample
### SDK功能
- 支持8416种菜品的分类识别,并给出置信度。
- 提供两个可用模型例子
1). 大模型(resnet50)例子:DishesClassificationExample
2). 小模型(mobilenet_v2)例子:LightDishesClassExample

[Dish Classification](https://aias-home.oss-cn-beijing.aliyuncs.com/AIAS/dish_sdk/dishes.txt)
[菜品分类](https://aias-home.oss-cn-beijing.aliyuncs.com/AIAS/dish_sdk/dishes.txt)

### Running Examples
-Test image
![dish](https://aias-home.oss-cn-beijing.aliyuncs.com/AIAS/dish_sdk/dish.jpeg)
### 运行例子
- 测试图片
![dish](https://aias-home.oss-cn-beijing.aliyuncs.com/AIAS/dish_sdk/dish.jpeg)

After successful execution, the command line should display the following information:
运行成功后,命令行应该看到下面的信息:
```text
清炒虾仁 : 1.0
[INFO ] - [
Expand All @@ -31,10 +37,30 @@ After successful execution, the command line should display the following inform
]
```

### Open source algorithms
#### 1. Open source algorithms used in the SDK
### 开源算法
#### 1. sdk使用的开源算法
[PaddleClas](https://github.com/PaddlePaddle/PaddleClas/blob/release%2F2.2/README_ch.md)
#### 2. How to export models?
#### 2. 模型如何导出 ?
- [export_model](https://github.com/PaddlePaddle/PaddleClas/blob/release%2F2.2/tools/export_model.py)
- [how_to_create_paddlepaddle_model](http://docs.djl.ai/docs/paddlepaddle/how_to_create_paddlepaddle_model_zh.html)



### 其它帮助信息
https://aias.top/guides.html

### Git地址:
[Github链接](https://github.com/mymagicpower/AIAS)
[Gitee链接](https://gitee.com/mymagicpower/AIAS)


#### 帮助文档:
- https://aias.top/guides.html
- 1.性能优化常见问题:
- https://aias.top/AIAS/guides/performance.html
- 2.引擎配置(包括CPU,GPU在线自动加载,及本地配置):
- https://aias.top/AIAS/guides/engine_config.html
- 3.模型加载方式(在线自动加载,及本地配置):
- https://aias.top/AIAS/guides/load_model.html
- 4.Windows环境常见问题:
- https://aias.top/AIAS/guides/windows.html
8 changes: 7 additions & 1 deletion 1_image_sdks/classification/dish_sdk/README_cn.md
Original file line number Diff line number Diff line change
@@ -1,9 +1,15 @@
### 官网:
[官网链接](https://www.aias.top/)

### 下载模型,放置于models目录
### 下载模型
- 链接: https://pan.baidu.com/s/1No67j8xhXlDfx676P14yQw?pwd=gtdq

### 模型使用方法:
- 1. 用模型的名字搜索代码,找到模型的加载位置
- 2. 然后更新模型路径(代码里默认加载路径是:项目/models 文件夹)
- 3. 具体模型加载方法
- http://aias.top/AIAS/guides/load_model.html

### 菜品分类识别SDK
菜品识别sdk,支持8416种菜品的分类识别。

Expand Down
72 changes: 50 additions & 22 deletions 1_image_sdks/crowd_sdk/README.md
Original file line number Diff line number Diff line change
@@ -1,23 +1,33 @@
### 官网:
[官网链接](http://www.aias.top/)

### Download the model and place it in the /models directory
- Link: https://github.com/mymagicpower/AIAS/releases/download/apps/crowdnet.zip
### 下载模型
- 链接: https://pan.baidu.com/s/1aFIcYOE2uzlmQFWEJopBoQ?pwd=svvt

### Crowd Density Detection SDK
The CrowdNet model is a crowd density estimation model proposed in 2016. The paper is "CrowdNet: A Deep Convolutional Network for DenseCrowd Counting". The CrowdNet model is mainly composed of deep convolutional neural networks and shallow convolutional neural networks. It is trained by inputting the original image and the density map obtained by the Gaussian filter. Finally, the model estimates the number of people in the image. Of course, this can not only be used for crowd density estimation, theoretically, density estimation of other animals, etc. should also be possible.
### 模型使用方法:
- 1. 用模型的名字搜索代码,找到模型的加载位置
- 2. 然后更新模型路径(代码里默认加载路径是:项目/models 文件夹)
- 3. 具体模型加载方法
- http://aias.top/AIAS/guides/load_model.html

The following is the structural diagram of the CrowdNet model. From the structural diagram, it can be seen that the CrowdNet model is composed of a deep convolutional network (Deep Network) and a shallow convolutional network (Shallow Network). The two groups of networks are spliced into one network and then input into a convolutional layer with a convolutional kernel size of 1. Finally, a density map data is obtained through interpolation, and the estimated number of people can be obtained by counting this density.
### 人群密度检测 SDK
CrowdNet模型是2016年提出的人流密度估计模型,论文为《CrowdNet: A Deep Convolutional Network for DenseCrowd Counting》,
CrowdNet模型主要有深层卷积神经网络和浅层卷积神经组成,通过输入原始图像和高斯滤波器得到的密度图进行训练,最终得到的模型估计图像中的行人的数量。
当然这不仅仅可以用于人流密度估计,理论上其他的动物等等的密度估计应该也可以。

以下是CrowdNet模型的结构图,从结构图中可以看出,CrowdNet模型是深层卷积网络(Deep Network)和浅层卷积网络(Shallow Network)组成,
两组网络通过拼接成一个网络,接着输入到一个卷积核数量和大小都是1的卷积层,最后通过插值方式得到一个密度图数据,通过统计这个密度就可以得到估计人数。
![model](https://aias-home.oss-cn-beijing.aliyuncs.com/AIAS/sec_sdks/images/network.png)

### SDK functions:
- Calculate the number of people
- Calculate the density map
### sdk功能:
- 统计人数
- 计算密度图

### Running Example- CrowdDetectExample
- Test picture
![crowd](https://aias-home.oss-cn-beijing.aliyuncs.com/AIAS/sec_sdks/images/crowd1.jpg)
### 运行例子 - CrowdDetectExample
- 测试图片
![crowd](https://aias-home.oss-cn-beijing.aliyuncs.com/AIAS/sec_sdks/images/crowd1.jpg)

- Example code:
- 例子代码:
```text
Path imageFile = Paths.get("src/test/resources/crowd1.jpg");
Image image = ImageFactory.getInstance().fromFile(imageFile);
Expand All @@ -28,21 +38,22 @@ The following is the structural diagram of the CrowdNet model. From the structur
Predictor<Image, NDList> predictor = model.newPredictor()) {
NDList list = predictor.predict(image);
//person quantity
//quantity为人数
float q = list.get(1).toFloatArray()[0];
int quantity = (int)(Math.abs(q) + 0.5);
logger.info("人数 quantity: {}", quantity);
// density为密度图
NDArray densityArray = list.get(0);
logger.info("density: {}", densityArray.toDebugString(1000000000, 1000, 1000, 1000));
logger.info("密度图 density: {}", densityArray.toDebugString(1000000000, 1000, 1000, 1000));
```


- After the operation is successful, the command line should see the following information:
- 运行成功后,命令行应该看到下面的信息:
```text
[INFO ] - Person quantity: 11
[INFO ] - 人数 quantity: 11
[INFO ] - Density: ND: (1, 1, 80, 60) cpu() float32
[INFO ] - 密度图 density: ND: (1, 1, 80, 60) cpu() float32
[
[ 4.56512964e-04, 2.19504116e-04, 3.44428350e-04, ..., -1.44560239e-04, 1.58709008e-04],
[ 9.59073077e-05, 2.53924576e-04, 2.51444580e-04, ..., -1.64886122e-04, 1.14555296e-04],
Expand All @@ -56,14 +67,14 @@ The following is the structural diagram of the CrowdNet model. From the structur
]
```
#### Density map
#### 密度图
![density](https://aias-home.oss-cn-beijing.aliyuncs.com/AIAS/sec_sdks/images/density.png)


### Open source algorithm
#### 1. Open source algorithm used by SDK
### 开源算法
#### 1. sdk使用的开源算法
- [PaddlePaddle-CrowdNet](https://github.com/yeyupiaoling/PaddlePaddle-CrowdNet)
#### 2. How to export the model?
#### 2. 模型如何导出 ?
- [how_to_create_paddlepaddle_model](http://docs.djl.ai/docs/paddlepaddle/how_to_create_paddlepaddle_model_zh.html)
- export_model.py
```text
Expand Down Expand Up @@ -94,4 +105,21 @@ exe = fluid.Executor(place)
if __name__ == '__main__':
paddle.enable_static()
save_pretrained()
```
```


### Git地址:
[Github链接](https://github.com/mymagicpower/AIAS)
[Gitee链接](https://gitee.com/mymagicpower/AIAS)


#### 帮助文档:
- http://aias.top/guides.html
- 1.性能优化常见问题:
- http://aias.top/AIAS/guides/performance.html
- 2.引擎配置(包括CPU,GPU在线自动加载,及本地配置):
- http://aias.top/AIAS/guides/engine_config.html
- 3.模型加载方式(在线自动加载,及本地配置):
- http://aias.top/AIAS/guides/load_model.html
- 4.Windows环境常见问题:
- http://aias.top/AIAS/guides/windows.html
8 changes: 7 additions & 1 deletion 1_image_sdks/crowd_sdk/README_cn.md
Original file line number Diff line number Diff line change
@@ -1,9 +1,15 @@
### 官网:
[官网链接](http://www.aias.top/)

### 下载模型,放置于models目录
### 下载模型
- 链接: https://pan.baidu.com/s/1aFIcYOE2uzlmQFWEJopBoQ?pwd=svvt

### 模型使用方法:
- 1. 用模型的名字搜索代码,找到模型的加载位置
- 2. 然后更新模型路径(代码里默认加载路径是:项目/models 文件夹)
- 3. 具体模型加载方法
- http://aias.top/AIAS/guides/load_model.html

### 人群密度检测 SDK
CrowdNet模型是2016年提出的人流密度估计模型,论文为《CrowdNet: A Deep Convolutional Network for DenseCrowd Counting》,
CrowdNet模型主要有深层卷积神经网络和浅层卷积神经组成,通过输入原始图像和高斯滤波器得到的密度图进行训练,最终得到的模型估计图像中的行人的数量。
Expand Down
8 changes: 7 additions & 1 deletion 1_image_sdks/face_sdks/face_alignment_sdk/README.md
Original file line number Diff line number Diff line change
@@ -1,9 +1,15 @@
### 官网:
[官网链接](https://www.aias.top/)

### 下载模型,放置于models目录
### 下载模型
- 链接: https://pan.baidu.com/s/10lwRrO3puEMpnVHD9FYObg?pwd=c134

### 模型使用方法:
- 1. 用模型的名字搜索代码,找到模型的加载位置
- 2. 然后更新模型路径(代码里默认加载路径是:项目/models 文件夹)
- 3. 具体模型加载方法
- http://aias.top/AIAS/guides/load_model.html

### 人脸对齐 SDK
sdk给出了人脸对齐的 java 实现。
人脸对齐算法是一种用于对齐人脸图像中的人脸的技术。其目的是将人脸图像中的人脸调整到一个标准的位置和大小,以便于后续的人脸识别、表情识别、面部分析等任务的进行。
Expand Down
6 changes: 6 additions & 0 deletions 1_image_sdks/face_sdks/face_detection_sdk/README.md
Original file line number Diff line number Diff line change
Expand Up @@ -4,6 +4,12 @@
### 下载模型,放置于models目录
- 链接: https://pan.baidu.com/s/10lwRrO3puEMpnVHD9FYObg?pwd=c134

### 模型使用方法:
- 1. 用模型的名字搜索代码,找到模型的加载位置
- 2. 然后更新模型路径(代码里默认加载路径是:项目/models 文件夹)
- 3. 具体模型加载方法
- http://aias.top/AIAS/guides/load_model.html

#### 人脸检测(含5个人脸关键点)SDK
sdk给出了人脸检测的 java 实现。
人脸检测算法是一种计算机视觉技术,它基于机器学习和人工智能技术,用于自动检测和定位图像或视频中的人脸。这种算法可以识别脸部特征和形状,并根据这些信息对人脸进行分类和跟踪。
Expand Down
6 changes: 6 additions & 0 deletions 1_image_sdks/face_sdks/face_feature_ir101_sdk/README.md
Original file line number Diff line number Diff line change
Expand Up @@ -4,6 +4,12 @@
### 下载模型,放置于models目录
- 链接: https://pan.baidu.com/s/10lwRrO3puEMpnVHD9FYObg?pwd=c134

### 模型使用方法:
- 1. 用模型的名字搜索代码,找到模型的加载位置
- 2. 然后更新模型路径(代码里默认加载路径是:项目/models 文件夹)
- 3. 具体模型加载方法
- http://aias.top/AIAS/guides/load_model.html

### 人脸特征提取与比对SDK
人工智能人脸特征提取是一种复杂而重要的计算机技术,其主要目的是通过对人脸图像进行深度分析和处理,提取出具有代表性的特征信息,以用于实现人脸识别、人脸比对、人脸验证等应用。这个技术的核心在于将人脸图像转化为计算机能够理解的数字特征,这些特征可以被用于训练人工智能模型,从而提高模型的准确性和性能。
人脸特征提取技术是一项非常重要的技术,在现代生活中被广泛应用于安防、金融、医疗等领域。在安防领域,人脸特征提取技术可以用于实现人脸识别、身份验证等功能,提高社会安全。在金融领域,人脸特征提取技术可以用于实现客户身份验证和授权,以确保金融交易的安全性和可靠性。在医疗领域,人脸特征提取技术可以用于实现医生和患者身份认证,提高医疗服务的质量和效率。
Expand Down
6 changes: 6 additions & 0 deletions 1_image_sdks/face_sdks/face_feature_sdk/README.md
Original file line number Diff line number Diff line change
Expand Up @@ -4,6 +4,12 @@
### 下载模型,放置于models目录
- 链接: https://pan.baidu.com/s/10lwRrO3puEMpnVHD9FYObg?pwd=c134

### 模型使用方法:
- 1. 用模型的名字搜索代码,找到模型的加载位置
- 2. 然后更新模型路径(代码里默认加载路径是:项目/models 文件夹)
- 3. 具体模型加载方法
- http://aias.top/AIAS/guides/load_model.html

### 人脸特征提取与比对SDK
人工智能人脸特征提取是一种复杂而重要的计算机技术,其主要目的是通过对人脸图像进行深度分析和处理,提取出具有代表性的特征信息,以用于实现人脸识别、人脸比对、人脸验证等应用。这个技术的核心在于将人脸图像转化为计算机能够理解的数字特征,这些特征可以被用于训练人工智能模型,从而提高模型的准确性和性能。
人脸特征提取技术是一项非常重要的技术,在现代生活中被广泛应用于安防、金融、医疗等领域。在安防领域,人脸特征提取技术可以用于实现人脸识别、身份验证等功能,提高社会安全。在金融领域,人脸特征提取技术可以用于实现客户身份验证和授权,以确保金融交易的安全性和可靠性。在医疗领域,人脸特征提取技术可以用于实现医生和患者身份认证,提高医疗服务的质量和效率。
Expand Down
Loading

0 comments on commit 458314a

Please sign in to comment.