Artificial Intelligence and computer vision are becoming an important part of modern embedded systems. Tasks that once required a Raspberry Pi or a desktop computer can now be performed on compact microcontroller boards. The HiWonder ESP32-S3 AI Vision Camera is one such development board that combines Wi-Fi, Bluetooth, an ESP32-S3 processor, PSRAM, and a camera module into a single platform.
This board is suitable for a wide range of applications including face detection, face recognition, colour tracking, QR code recognition, object classification, and smart robotics. Since it is based on the ESP32-S3, it can also be programmed using the familiar Arduino IDE, making it accessible to beginners as well as experienced developers.
If you have just purchased the board and are wondering where to begin, this guide will walk you through the complete setup process. By the end of this tutorial, you will have the Arduino IDE configured correctly, the required ESP32 board package installed, and a live video stream running in your web browser.
Before You Begin
You will need the following hardware and software before proceeding.
Hardware
- HiWonder ESP32-S3 AI Vision Camera
- USB Type-C data cable
- Computer running Windows, Linux, or macOS
- 2.4 GHz Wi-Fi network
Software
- Arduino IDE 2.x
- ESP32 Board Package version 3.0.7
At the time of writing, most of the AI examples provided for this camera were developed using ESP32 Arduino Core 3.0.7. Although newer versions of the ESP32 board package are available, several AI libraries used by the examples are not compatible with the latest releases. Installing version 3.0.7 avoids compilation errors and provides the best compatibility with the supplied examples.
Installing the ESP32 Board Package
If this is your first time working with an ESP32 board, Arduino IDE needs to be configured to download the required board definitions.
Open Arduino IDE and navigate to File → Preferences. Locate the field named Additional Boards Manager URLs and add the following address.
https://raw.githubusercontent.com/espressif/arduino-esp32/gh-pages/package_esp32_index.json
If another URL is already present, simply separate the entries with a comma.
After saving the preferences, open Tools → Board → Boards Manager. Search for ESP32 and locate ESP32 by Espressif Systems.
Instead of installing the latest version, open the version drop-down menu and select 3.0.7. Click Install and wait for the installation to complete. Depending on your internet connection, this may take a few minutes.
Once the installation finishes, restart Arduino IDE. Restarting ensures that all board definitions and tools are loaded correctly.
Selecting the Correct Board
Reconnect the camera board if necessary and select the appropriate COM port from Tools → Port.
Next, navigate to Tools → Board and choose ESP32S3 Dev Module.
Several board options are available for the ESP32-S3 family. The HiWonder AI Vision Camera works correctly with the generic ESP32S3 Dev Module, so there is no need to search for a separate board definition.
Recommended Board Configuration
Before uploading any program, configure the following settings from the Tools menu.
| Setting | Value |
|---|---|
| Board | ESP32S3 Dev Module |
| Flash Size | 16 MB |
| PSRAM | OPI PSRAM |
| CPU Frequency | 240 MHz |
| USB CDC On Boot | Enabled |
| Upload Speed | 921600 |
| Partition Scheme | 16M Flash (3MB APP / 9.9MB FATFS) |
These settings have been tested with the HiWonder ESP32-S3 AI Vision Camera and provide sufficient memory for both camera streaming and AI applications.
Uploading the Web Streaming Example
With the development environment ready, the next step is to upload a simple web streaming application. This is the easiest way to verify that the camera hardware is functioning correctly.
Open the provided web streaming sketch in Arduino IDE. Before compiling the program, replace the default Wi-Fi credentials with your own network name and password.
const char* ssid = "YOUR_WIFI_NAME";
const char* password = "YOUR_WIFI_PASSWORD";
Click Verify to compile the sketch. If the compilation completes without any errors, click Upload. The firmware will be built and transferred to the ESP32-S3 board.
After the upload is complete, open the Serial Monitor and set the baud rate to 115200. Within a few seconds, the board should connect to your Wi-Fi network and print its assigned IP address.
Copy this IP address and open it in any web browser connected to the same network. If everything has been configured correctly, you should see a live video stream from the camera.
Reaching this stage confirms that the hardware, camera interface, Wi-Fi connection, and software configuration are all working as expected. It also provides a solid foundation for exploring more advanced AI examples in the next tutorials.
What’s Next?
Once the live stream is working, you are ready to move on to the AI capabilities of the board. In the upcoming articles, we will cover face detection, face recognition, colour detection, QR code recognition, object classification, and building custom AI applications using the ESP32-S3 AI Vision Camera.
