optimize docs

README.md

@@ -28,14 +28,20 @@ English | [中文](./README_ZH.md)
 
 </div>
 
+![MaixCAM](https://wiki.sipeed.com/maixpy/static/image/maixcams.png)
 
-> New MaixPy (v4) and new hardware platform MaixCAM is coming now!<br>
-> If you have any suggestions, tell us on [MaixHub](https://maixhub.com/discussion/100073), or [Telegram/MaixPy](https://t.me/maixpy) or QQ group: 862340358.
->
-> Click the **Star** in the upper right corner to let us know you like it to encourage us to add more features.
 
+## Feature Overview
 
-## Introduction
+MaixPy offers simple Python programming combined with powerful edge computing hardware. Integrated hardware peripheral operations, video streaming, AI vision algorithms, and audio algorithms etc. With its plug-and-play design, MaixPy enables you to quickly implement your intelligent projects.
+
+Additionally, MaixPy provides the MaixVision IDE, MaixHub online training platform, detailed documentation, and even a C/C++ SDK with identical APIs, ensuring seamless development and production deployment.
+
+For more features and video demonstrations, please visit the **official website: [wiki.sipeed.com/maixpy/](https://wiki.sipeed.com/maixpy/)**
+
+> If you like this project, please click **Star** on the top right of the [MaixPy Project](https://github.com/sipeed/maixpy) to encourage us to develop more exciting content!
+
+## Concise and Efficient Code (API) Design
 
 With MaixPy you can easily create **AI vision project** within **10 lines of code**:
 
@@ -55,13 +61,12 @@ while 1:
     disp.show(img)
 ```
 
-Result video:
+Result:
+![](https://wiki.sipeed.com/maixpy/static/video/classifier.gif)
 
-<video playsinline controls autoplay loop muted preload src="https://github.com/sipeed/MaixPy/assets/8625829/d06e4f0e-31af-4679-b3bc-1f93cd633af0" type="video/mp4">
-Classifier Result video
-</video>
+## Edge(embeded) friendly
 
-Simply use **peripheral** like serial port:
+Simply use hardware **peripheral** like serial port:
 
 ```python
 from maix import uart
@@ -73,39 +78,35 @@ serial.write_str("hello world")
 print("received:", serial.read(timeout = 2000))
 ```
 
+## MaixVision workstation
+
 We also provide a handy **[MaixVision](https://wiki.sipeed.com/en/maixvision)** workstation software to make development easier and faster:
 
 <video playsinline controls muted preload src="https://github.com/sipeed/MaixPy/assets/8625829/1168a282-d7c2-45bc-9ffb-c00de1ca24f5" type="video/mp4">
 MaixVision
 </video>
 
-And online AI train platform **[MaixHub](https://maixhub.com)**, one click to train AI model and deploy to MaixCAM even you have no AI knowledge and expensive training equipment.
-
-![MaixHub](https://wiki.sipeed.com/maixpy/static/image/maixhub.jpg)
+## MaixHub online platform
 
-## Features
+**[MaixHub](https://maixhub.com)** provide free online AI train service, one click to train AI model and deploy to MaixCAM even you have no AI knowledge and expensive training equipment.
 
-Python programing, MaixVision Workstation, AI vision, video streaming, voice recognize, peripheral usage etc.
+![MaixHub](https://wiki.sipeed.com/maixpy/static/image/maixhub.jpg)
 
-**Details and videos** visit official site: [wiki.sipeed.com/maixpy/](https://wiki.sipeed.com/maixpy/)
 
 ## Hardware platform MaixCAM
 
-And we provide new powerful hardware platform **[MaixCAM](https://wiki.sipeed.com/maixcam)** and **[MaixCAM-Pro](https://wiki.sipeed.com/maixcam-pro)**:
+And we provide two powerful hardware platform **[MaixCAM](https://wiki.sipeed.com/maixcam)** and **[MaixCAM-Pro](https://wiki.sipeed.com/maixcam-pro)**, with datasheet register level open.
 
-![MaixCAM](https://wiki.sipeed.com/maixpy/static/image/maixcam_pro.png)
+![MaixCAM](https://wiki.sipeed.com/maixpy/static/image/maixcams.png)
 
 | CPU | NPU | Memory |
 | --- | --- | ------- |
-| - 1GHz RISC-V(Linux)/ARM A53<br>- 700MHz RISCV-V(RTOS)<br>- 25~300MHz 8051(LowPower) | 1Tops@INT8 NPU, support BF16<br>support YOLOv5 YOLOv8 etc.| 256MB DDR3 |
+| - 1GHz RISC-V(Linux)<br>- 700MHz RISCV-V(RTOS)<br>- 25~300MHz 8051(LowPower) | 1Tops@INT8 NPU, support BF16<br>support YOLOv5 YOLOv8 etc.| 256MB DDR3 |
 
 | Connecting | Peripheral | MultiMedia | Buy |
 | ----------- | ----- | --- | ---- |
 |  USB2.0/WiFi6/BLE5.4 | IIC/PWM/SPI/UART/WDT/ADC | - 4M Camera<br>- 2.3" 552x368 Touchscreen<br>- H.264/H.265/MJPEG codec | [Sipeed Official Store](https://wiki.sipeed.com/store) |
 
-Chip register level open, more detalils: [MaixCAM-Pro](https://wiki.sipeed.com/maixcam-pro) or [MaixCAM](https://wiki.sipeed.com/maixcam)
-> * **Maix-I K210** series is outdated, MaixPy v4 not support it, use it please visit **[MaixPy-v1](https://github.com/sipeed/maixpy-v1)**
-
 
 ## Who are using MaixPy?
 
@@ -186,9 +187,11 @@ All files in this repository are under the terms of the [Apache License 2.0 Sipe
 ## Community
 
 * Project sharing: [maixhub.com/share](https://maixhub.com/share)
+* App sharing: [maixhub.com/app](https://maixhub.com/app)
 * Discussion: [maixhub.com/discussion](https://maixhub.com/discussion)
 * QQ group: 862340358
 * Telegram: [t.me/maixpy](https://t.me/maixpy)
+* Github issues: [github.com/sipeed/maixpy/issues](https://github.com/sipeed/maixpy/issues)
 
 
 

README_ZH.md

@@ -13,7 +13,7 @@ MaixPy (v4)
     <a href="https://wiki.sipeed.com/maixpy/doc/index.html"> 快速开始 </a> |
     <a href="https://wiki.sipeed.com/maixpy/index.html"> 文档 </a> |
     <a href="https://wiki.sipeed.com/maixpy/api/index.html"> API </a> |
-    <a href="https://wiki.sipeed.com/maixcam"> 硬件 </a>
+    <a href="https://wiki.sipeed.com/maixcam-pro"> 硬件 </a>
 </h3>
 
 [![GitHub Repo stars](https://img.shields.io/github/stars/sipeed/MaixPy?style=social)](https://github.com/sipeed/MaixPy/stargazers)
@@ -29,13 +29,20 @@ MaixPy (v4)
 </div>
 
 
+![MaixCAM](https://wiki.sipeed.com/maixpy/static/image/maixcams.png)
 
-> 新的 MaixPy (v4) 和新的硬件平台 MaixCAM 来了!<br>
-> 如果你有任何建议，请在 [MaixHub](https://maixhub.com/discussion/100073) 上告诉我们，或者 [Telegram/MaixPy](https://t.me/maixpy) 或 QQ 群: 862340358。
->
-> 点击右上角的 **Star** 让我们知道你喜欢它，以鼓励我们开发更多功能。
 
-## 介绍
+## 特性简介
+
+MaixPy 提供简易的 Python 编程和性能强大的边缘计算硬件，内置了大量易用的 硬件外设操作、视频串流、AI 视觉算法、听觉算法，开箱即用，帮助你快速落地你的智能化项目。
+
+并且提供 MaixVision IDE, MaixHub 云端训练平台，详细的文档，甚至有相同 API 的 C/C++ SDK，帮助你无障碍开发和量产落地。
+
+更多特性和视频展示请访问**官网: [wiki.sipeed.com/maixpy/](https://wiki.sipeed.com/maixpy/)**
+
+> 喜欢请点击 [MaixPy 项目](https://github.com/sipeed/maixpy)右上角 **Star** 鼓励我们开发更多有趣内容！
+
+## 简洁高效的代码(API)设计
 
 使用 MaixPy 轻松创建 **AI 视觉项目**，只需 **10 行代码**:
 
@@ -55,13 +62,12 @@ while 1:
     disp.show(img)
 ```
 
-效果视频:
+效果:
+![](https://wiki.sipeed.com/maixpy/static/video/classifier.gif)
 
-<video playsinline controls autoplay loop muted preload src="https://github.com/sipeed/MaixPy/assets/8625829/d06e4f0e-31af-4679-b3bc-1f93cd633af0" type="video/mp4">
-Classifier Result video
-</video>
+## 边缘计算设备（嵌入式）友好
 
-建议的代码使用串口等 **外设**：
+运行在边缘计算设备，不光有视觉相关，常用的外设也不在话下，比如使用串口：
 
 ```python
 from maix import uart
@@ -73,39 +79,34 @@ serial.write_str("hello world")
 print("received:", serial.read(timeout = 2000))
 ```
 
-我们还提供了便捷的 **[MaixVision](https://wiki.sipeed.com/maixvision)** 工作站，让开发更简单快速：
+## MaixVision 工作站
+
+提供便捷的 **[MaixVision](https://wiki.sipeed.com/maixvision)** 工作站（IDE），让开发更简单快速：
 
 <video playsinline controls muted preload src="https://github.com/sipeed/MaixPy/assets/8625829/1168a282-d7c2-45bc-9ffb-c00de1ca24f5" type="video/mp4">
 MaixVision
 </video>
 
-以及在线 AI 训练平台 **[MaixHub](https://maixhub.com)**， 无需 AI 基础和昂贵的训练设备，一键训练模型，一键部署到 MaixCAM。
-
-![MaixHub](https://wiki.sipeed.com/maixpy/static/image/maixhub.jpg)
-
-## 功能特点
-
-Python 编程, MaixVision 工作站, AI 视觉, 视频串流, 语音识别, 外设使用 等。
+## MaixHub 在线平台
 
-**细节和视频展示** 请访问官网: [wiki.sipeed.com/maixpy/](https://wiki.sipeed.com/maixpy/)
+**[MaixHub](https://maixhub.com)** 提供免费在线 AI 训练， 无需 AI 基础和昂贵的训练设备，一键训练模型，一键部署。
 
+![MaixHub](https://wiki.sipeed.com/maixpy/static/image/maixhub.jpg)
 
-## 硬件平台 MaixCAM
+## 性能强劲的硬件平台
 
-提供性能强劲的硬件平台 **[MaixCAM](https://wiki.sipeed.com/maixcam)** 和 **[MaixCAM-Pro](https://wiki.sipeed.com/maixcam-pro)**:
+目前支持两款硬件平台， 芯片寄存器级别开放，详细请看 **[MaixCAM](https://wiki.sipeed.com/maixcam)** 和 **[MaixCAM-Pro](https://wiki.sipeed.com/maixcam-pro)**:
 
-![MaixCAM](https://wiki.sipeed.com/maixpy/static/image/maixcam-pro.png)
+![MaixCAM](https://wiki.sipeed.com/maixpy/static/image/maixcams.png)
 
 | CPU | NPU | Memory |
 | --- | --- | ------- |
-| - 1GHz RISC-V(Linux)/ARM A53<br>- 700MHz RISCV-V(RTOS)<br>- 25~300MHz 8051(LowPower) | 1Tops@INT8 NPU, support BF16<br>support YOLOv5 YOLOv8 etc.| 256MB DDR3 |
+| - 1GHz RISC-V(Linux)<br>- 700MHz RISCV-V(RTOS)<br>- 25~300MHz 8051(LowPower) | 1Tops@INT8 NPU, support BF16<br>support YOLOv5 YOLOv8 etc.| 256MB DDR3 |
 
-| Connecting | Peripheral | MultiMedia | Price |
-| ----------- | ----- | --- | ---- |
-|  USB2.0/WiFi6/BLE5.4 | IIC/PWM/SPI/UART/WDT/ADC | - 4M Camera<br>- 2.3" 552x368 Touchscreen<br>- H.264/H.265/MJPEG codec | [￥169/￥249](https://wiki.sipeed.com/store) |
+| 连接 | 外设 | 多媒体 | 购买 |
+| ----------- | ----- | --- | --- |
+|  USB2.0/WiFi6/BLE5.4 | IIC/PWM/SPI/UART/WDT/ADC | - 4M Camera<br>- 2.3" 552x368 Touchscreen<br>- H.264/H.265/MJPEG codec | [Sipeed 官方店铺](https://wiki.sipeed.com/store) |
 
-芯片寄存器级别开放，详情看 [MaixCAM-Pro](https://wiki.sipeed.com/maixcam-pro) 或 [MaixCAM](https://wiki.sipeed.com/maixcam)
-> * **Maix-I K210** 已过时，MaixPy v4 不支持它, 仍要使用请访问 **[MaixPy-v1](https://github.com/sipeed/maixpy-v1)**
 
 
 ## 谁在用 MaixPy？
@@ -188,9 +189,11 @@ MaixPy 不仅仅是一个 Python SDK，还有一个完整的生态系统，包
 ## 社区
 
 * 项目分享: [maixhub.com/share](https://maixhub.com/share)
+* 应用分享: [maixhub.com/app](https://maixhub.com/app)
 * 讨论: [maixhub.com/discussion](https://maixhub.com/discussion)
 * QQ 群: 862340358
 * Telegram: [t.me/maixpy](https://t.me/maixpy)
+* Github issues: [github.com/sipeed/maixpy/issues](https://github.com/sipeed/maixpy/issues)
 
 
 

docs/doc/en/ai_model_converter/maixcam.md

@@ -46,6 +46,13 @@ For example, in `YOLOv5`:
 
 There are three `conv` layers, with subsequent calculations handled by the CPU. For quantization, use the outputs of these `conv` layers as the final outputs of the model. The output names in this case are `/model.24/m.0/Conv_output_0,/model.24/m.1/Conv_output_0,/model.24/m.2/Conv_output_0`.
 
+YOLO11/YOLOv8: Please refer to [Offline Training YOLO11/YOLOv8](../vision/customize_model_yolov8.md).
+
+For classification models, it is generally sufficient to take the name of the last output layer. However, if there is a `softmax` layer, it is recommended not to include `softmax` in the model. Instead, take the output of the layer before `softmax`. In the diagram below, there is no `softmax` layer, so the final layer can be used directly.
+
+![](../../assets/mobilenet_top.png)
+
+
 ## Setting Up the Model Conversion Environment
 
 The model conversion uses Sophgo's [https://github.com/sophgo/tpu-mlir](https://github.com/sophgo/tpu-mlir). We will install it in a Docker environment to avoid compatibility issues with the host machine.

docs/doc/en/sidebar.yaml

@@ -83,14 +83,14 @@ items:
         label: Detect with angle(OBB)
     -   file: vision/hand_landmarks.md
         label: Hand landmarks
+    -   file: vision/body_pose_classification.md
+        label: Human pose classifier
     -   file: vision/maixhub_train.md
         label: MaixHub online AI training
     -   file: vision/customize_model_yolov5.md
         label: YOLOv5 model offline training
     -   file: vision/customize_model_yolov8.md
         label: YOLO11/v8 model offline training
-    -   file: vision/body_pose_classification.md
-        label: YOLO11 pose model based classifier
     -   file: ai_model_converter/maixcam.md
         label: ONNX model to MaixCAM's
     -   file: pro/datasets.md

docs/doc/en/vision/customize_model_yolov8.md

@@ -94,8 +94,15 @@ Pay attention to the selection of the model output nodes (note that the numerica
 * Image segmentation:
   * YOLOv8-seg extracts `/model.22/dfl/conv/Conv_output_0,/model.22/Sigmoid_output_0,/model.22/Concat_output_0,output1`.
   * YOLO11-seg extracts `/model.23/dfl/conv/Conv_output_0,/model.23/Sigmoid_output_0,/model.23/Concat_output_0,output1`.
+* OBB Detection:
+  * YOLOv8 extracts`/model.22/dfl/conv/Conv_output_0,/model.22/Sigmoid_1_output_0,/model.22/Sigmoid_output_0`as outputs.
+  * YOLO11 extracts`/model.23/dfl/conv/Conv_output_0,/model.23/Sigmoid_1_output_0,/model.23/Sigmoid_output_0`as outputs.
 
-![](../../assets/yolov8_out1.jpg) ![](../../assets/yolov8_out2.jpg)
+YOLOv8/YOLO11 output nodes:
+![](../../assets/yolov8_out.jpg)
+
+YOLOv8/YOLO11 OBB output nodes:
+[](../../assets/yolo11_out_obb.jpg)
 
 For object detection, the MUD file would be as follows (replace `yolo11` for YOLO11):
 

docs/doc/zh/ai_model_converter/maixcam.md

@@ -50,6 +50,11 @@ labels = person, bicycle, car, motorcycle, airplane, bus, train, truck, boat, tr
 
 可以看到这里有三个`conv`，后面的计算均由 CPU 进行，我们量化时就采取这几个`conv`的输出作为模型的最后输出，在这里输出名分别叫`/model.24/m.0/Conv_output_0,/model.24/m.1/Conv_output_0,/model.24/m.2/Conv_output_0`。
 
+YOLO11/YOLOv8 请看[离线训练 YOLO11/YOLOv8](../vision/customize_model_yolov8.md).
+
+分类模型一般来说取最后一个输出名称就行，不过如果有`osftmax`的话，建议不把`softmax`包含在模型里面，即取`softmax`前一层的输出名，下图是没有`softmax`层的所以直接取最后一层即可。
+![](../../assets/mobilenet_top.png)
+
 
 ## 安装模型转换环境
 

docs/doc/zh/sidebar.yaml

@@ -83,14 +83,14 @@ items:
         label: 带旋转角度的检测(OBB)
     -   file: vision/hand_landmarks.md
         label: 手部关键点检测
+    -   file: vision/body_pose_classification.md
+        label: 人体姿态分类器
     -   file: vision/maixhub_train.md
         label: MaixHub 在线训练 AI 模型
     -   file: vision/customize_model_yolov5.md
         label: 离线训练 YOLOv5 模型
     -   file: vision/customize_model_yolov8.md
         label: 离线训练 YOLO11/YOLOv8 模型
-    -   file: vision/body_pose_classification.md
-        label: 基于 YOLO11 pose 模型的人体姿态分类器
     -   file: ai_model_converter/maixcam.md
         label: ONNX 模型转给 MaixCAM 用
     -   file: pro/datasets.md

docs/doc/zh/vision/customize_model_yolov8.md

@@ -96,7 +96,11 @@ MaixPy/MaixCDK 目前支持了 YOLOv8 / YOLO11 检测 以及 YOLOv8-pose / YOLO1
   * YOLOv8 提取`/model.22/dfl/conv/Conv_output_0,/model.22/Sigmoid_1_output_0,/model.22/Sigmoid_output_0`这三个输出。
   * YOLO11 提取`/model.23/dfl/conv/Conv_output_0,/model.23/Sigmoid_1_output_0,/model.23/Sigmoid_output_0`这三个输出。
 
-![](../../assets/yolov8_out1.jpg) ![](../../assets/yolov8_out2.jpg) ![](../../assets/yolo11_out_obb.jpg)
+YOLOv8/YOLO11 输出节点:
+![](../../assets/yolov8_out.jpg)
+
+YOLOv8/YOLO11 OBB 输出节点：
+![](../../assets/yolo11_out_obb.jpg)
 
 对于物体检测，mud 文件为（YOLO11 model_type 改为 yolo11）
 ```ini

docs/pages/index/README.md

@@ -174,7 +174,7 @@ h1 {
 
 <div class="mt-16"></div>
 
-<img class="text-center" src="/static/image/maixcam_pro.png">
+<img class="text-center" src="/static/image/maixcams.png">
 
 <div class="mt-6"></div>
 

docs/pages/index_en/README.md

@@ -173,7 +173,7 @@ English | [中文](../)
 
 <div class="mt-16"></div>
 
-<img class="text-center" src="/static/image/maixcam_pro.png">
+<img class="text-center" src="/static/image/maixcams.png">
 
 <div class="mt-6"></div>