propanedragon / pihueentertainment Goto Github PK

There's nothing in place to allow for automation of the application. This will probably require settings to be saved, so depends on #2 to be completed first.

At the moment there's no provisions for rotating the camera. This is critical for most functionality.

I've installed Qt on a Pi 3 by using

sudo apt-get install qt5-default
sudo apt-get install qtcreator

changed permission of the executable
sudo chmod +x ./HueEntertainmentCenter.out

./HueEntertainmentCenter.out

It can't detect the camera. I have tried both logitech webcam c920 and c270 webcams.
Right at the beginning I get a warning libEGL DRI2 failed to authenticate

image here (https://photos.app.goo.gl/5etL464TDuHZwTdV8)

There's no way of setting a maximum or minimum brightness setting at the moment so lights can get a bit overpowering. This also relies on #2.

Camera selection is currently based off of the default camera, which is a bit vague and doesn't really make sense. Make a dialog (like the entertainment area selection dialog) that allows users to choose between different connected cameras if there's more than one.

Currently the image that is worked on is only 50x50 pixels, which is fine for the most part, but it should be changeable to allow it to work better on less or more powerful hardware. This relies on #2 to save the image size.

The target framerate is currently limited to 10FPS. It would be nice to allow the user to change this setting to their own preferences. Requires #2 first.

There's no way of checking for updates at the moment, so features could go unnoticed.

The following features could be introduced:

• Check the TV activity and if there doesn't appear to be anything going on then set the lights back to their previous state and enter the low power mode.
• Allow for a timer to be set where the lights will be active. Anything outside this timer will set the lights to their previous state and enter low power mode.

This would allow the application to be used somewhat headless, without any interaction required.

Currently areas that change lights are hard coded to the left, right and centre of the camera capture. It would be nice if the user could modify these to suit their needs (such as capturing a smaller segment, or an offset from the defaults).

Selected settings should be stored somewhere so they don't have to be re-entered every time the application is started. At the moment there's no state stored, so the program has no memory of the previous settings.

Compared to the Windows camera application, this seems to run quite a bit behind it. I'm not sure whether some Qt Camera settings could be tweaked to give more performance but it seems to have little effect during experimentation. This needs some further investigation.