ESP-WHO [中文]
ESP-WHO is an image processing development platform based on Espressif chips. It contains development examples that may be applied in practical applications.
ESP-WHO provides examples such as Human Face Detection, Human Face Recognition, Pedestrian Detection etc. You can develop a variety of practical applications based on these examples. ESP-WHO runs on ESP-IDF. ESP-DL provides rich deep learning related interfaces for ESP-WHO, which can be implemented with various peripherals to realize many interesting applications.
- The repository has been fully refactored. It adapts to the new ESP-DL
- New chip ESP32-P4 is now supported.
- Camera and the deep learning model now runs asynchronously, which achieves higher fps.
- Add lvgl(Light and Versatile Graphics Library) support, feel free to develop your own graphical applications.
- New pedestrian detect model is added.
Some chip such as esp32 and esp32-s2, and examples such as cat face detection, color detection, code recognition is not available in this branch currently, we're still working on them. Old branch can be found here. old ESP-WHO branch
We recommend novice developers to use the development boards designed by Espressif. The examples provided by ESP-WHO are developed based on the following Espressif development board, and the corresponding relationships between the development boards and SoC are shown in the table below.
| Soc | ESP32-S3 | ESP32-P4 |
|---|---|---|
| Development Board | ESP-S3-EYE | ESP32-P4-Function-EV-Board |
Using a development board not mentioned in the table above, configure pins assigned to peripherals manually, such as camera, LCD, and buttons.
ESP-WHO runs on ESP-IDF release/v5.4 branch . For details on getting ESP-IDF, please refer to ESP-IDF Programming Guide.
Run the following commands in your terminal to download ESP-WHO:
git clone https://github.com/espressif/esp-who.gitAll examples of ESP-WHO are stored in examples folder. Structure of this folder is shown below:
├── examples
│ ├── human_face_detect // high fps with esp_lcd only
│ ├── human_face_detect_lvgl // slower but with more powerful graphic lib
│ ├── human_face_recognition // interact with button(s3) or touch screen(p4)
│ │ // to experience the face recognition demo.
│ ├── pedestrian_detect
│ └── pedestrian_detect_lvgl For the development boards mentioned in Hardware, all examples are available out of the box. To run the examples, you only need to perform Step 1, Step 2 and Step 3.
If you are using ESP32-P4-Function-EV-Board, please following the user guide ESP32-P4 user guide to connect the camera and lcd to the development board.
Open the terminal and go to any folder that stores examples (e.g. examples/human_face_detection/lcd). Run the following command to set the target chip:
idf.py set-target [SoC]Replace [SoC] with your target chip, e.g. esp32s3, esp32p4
In addtion to the default configuration, there may be some options in the example that you can modify freely. For more details, read the README.md under the examples.
idf.py menuconfigFlash the program and launch IDF Monitor:
idf.py flash monitorPlease submit an issue if you find any problems using our products, and we will reply as soon as possible.
