SERINDA v2

Osiyo. Dohiju? Hey welcome back.

This weekend I managed to accomplish a few items on my list. The biggest of which was encapsulating the camera startup code so I could display multiple cameras at the same time. There is some work to do in order to display those cameras dynamically. Meaning, right now you have to say how many cameras and have a url to hit. I’ll be setting it up so that you can hit one url and pass in the camera number and then it’ll do the rest.

My next steps are to get the same code to work on the Raspberry Pi. Then I’m going to add object recognition and face recognition. After that, I need to set it up so I can apply a filter to only one camera at a time instead of both cameras. So I could say “motion detect camera 1” where right now “detect motion” will apply to both cameras because they’re looping the same code. I’m going to add streaming OCR via Tesseract. I currently have TesseractJs on the page so you can upload an image and it’ll OCR. I need to set up the display so it can be adjusted by voice. Commands like move video 1 20 pixels to the left, etc. And then save the layout so it’ll be the same each time. The final item I’m going to add before I release the code publicly is gesture recognition. One camera is the integrated laptop camera (and eventually PiCam). The second camera is one I wear on a helmet. This second camera is the one I want to recognize gestures.

I’m very happy with the progress. I have a bunch of plugins I’d like to write from the past. Like the PDF viewer works, but the scrolling isn’t working correctly yet. I want to be able to say “scroll slow” and it’ll scroll however fast I need it to so I can read. There are other commands for scaling, rotating, scroll down x lines, etc. As well, I’d like to be able to use voice commands to load a PDF. I’d like to add some integration with the OS so I can open up programs and interact with them (pywinauto or some other framework). After that, I want to be able to interact with programs like Evernote for CRUD of notes – adding camera captures to notes, that type of thing. I would also like to add some AR/MX to the view. This also means eventually doing the display so I can see it in AR view. I know this is going to be a little bit harder to accomplish, but it’ll be worth it. At some point, I’d like to set it up to interface with google api and alexa and maybe more. Also at some point, I’ll offload work to the LattePanda if the RaspberryPi doesn’t quite work the way I need it to.

A lot of cool stuff coming up. I’m pretty excited about it.

Until next time. Dodadagohvi.