The ESP32-Cam is a fantastic way to add video streaming at low cost. 

However the example video application is somewhat unstable: the video stream most often stops after only a few seconds of streaming. I thus had to rewrite the video server.

Based on the experience gained with websockets on the ESP8266 main controller, I used a websocket server to send JPEG pictures to the client, and a little bit of javascript to update the HTML page each time a frame is received. 

The video stream is much more reliable this way.

Next step: I plan on using Posenet (running in the browser through tf.js, not on the ESP32 !) to identify people in front of Wall-E and enable support some physical interaction with them, e.g. turn head towards their face, use gestures to tell Wall-E to move, etc.

https://github.com/DIYEmbeddedSystems/wall-e-cam