ESP-WHO is a face detection and recognition platform that is currently based on Espressif Systems' [ESP32](https://espressif.com/en/products/hardware/esp32/overview) chip.
# Overview
ESP-WHO supports development of face detection and recognition applications based around Espressif Systems' [ESP32](https://espressif.com/en/products/hardware/esp32/overview) chip in the most convenient way. With ESP-WHO, you can easily build up face detection- and recognition-featured applications, for instance:
* A coffee machine that brews coffee according to your taste preference;
* Home applicance that will shut off the electricity automatically when unsupervised children are operating them;
* And other more applications that suit your needs.
In general, the ESP-WHO features will be supported as shown below:
![overview](img/overview.jpg)
In ESP-WHO, Detection, Recognition and Image Utility are at the core of the platform.
* **Detection** takes images as input and give the position of face if there is a face. It is implemented with MTMN model, which refers to [MTCNN](http://cn.arxiv.org/abs/1604.02878) and [MobileNets](https://arxiv.org/abs/1801.04381).
* **Optimization** is mainly to increase the precision of the inference, and to accelerate the whole process. But also it might change the structure of the network, update the coefficients, refactor the code, etc.
* Image sources could be **input** via camera. However, we don't provide many drivers right now, those for other camera modules will be released in the future.
* Results could be **output** and displayed through Command line, LCD or even website via Wi-Fi http service.
To run ESP-WHO, you need to have a development board which integrates a **ESP32 module** that has sufficient GPIO pins and more than **4 MB** external SPI RAM. Either [ESP-WROVER-KIT](https://www.espressif.com/en/products/hardware/esp-wrover-kit/overview) or [ESP-EYE](https://www.espressif.com/en/products/hardware/esp-eye/overview) can be a good choice as the test board.
If you clone project without `--recursive` flag, please go to the `esp-who` directory and run command `git submodule update --init` before doing anything.
> Now this repository supports two versions of IDF: v3.3.1 and v4.0.0. Master branch is based on v4.0.0, while for v3.3.1, please checkout to idfv3.3.1 branch.
The [esp-face](https://github.com/espressif/esp-face/tree/master) component contains the APIs of ESP-WHO neural networks, including face detection and recognition framework.
* [Check the Issues section on github](https://github.com/espressif/esp-who/issues) if you find a bug or have a feature request, please check existing Issues before opening a new one.
* If you're interested in contributing to ESP-WHO, please check the [Contributions Guide](https://esp-idf.readthedocs.io/en/latest/contribute/index.html).