If you put the kinect just infront of the glass will take up to 3-4meters... i suggest a treshold after the Chromakey. Or at least at the end so you will have pure black and white( 0-1) that works better with instances or blobtracking. Happy that the comunity or NERDS IS GROWING UP!!! LoL
Thank you for video! I honestly would like to see tutorial on that Rutt-Etra effect you are using :)) I recently did similar installation and it was pretty resource-demanding despite my render output was 840x840
hey! very cool tutorial, thank you so much! Just a little off topic question? How do you make the little green circle appear to change the values? :)))
hi :) thank u so much. if i wanna put new material to fill my the women in my video, how do i do that? or how i can put new background like u did in the begging of your example?
humm good point I haven't tested it with white T-Shirt 😅 You might get better results by caching more images at the beginning and / or playing with the strength of the edge TOP
I really like that approach, but unfortunately it only works with white skin and bright colored or white clothes. However you could use Math TOP's for limiting and normalizing instead of Level TOP's, and also adding some feedback at the end of the chain is improving this a little :)
If you put the kinect just infront of the glass will take up to 3-4meters... i suggest a treshold after the Chromakey. Or at least at the end so you will have pure black and white( 0-1) that works better with instances or blobtracking. Happy that the comunity or NERDS IS GROWING UP!!! LoL
I just started using my phone as a cam for TD via DroidCamX and this is great info, tyvm!
super cool! will give a try later today
Thank you for video! I honestly would like to see tutorial on that Rutt-Etra effect you are using :)) I recently did similar installation and it was pretty resource-demanding despite my render output was 840x840
did you start with a grid and increase point numbers until they were lines?
Awesome video really great info. Would it work better if I have a solid background (black/white/green screen) and contrast clothing?
yes! having a solid background would also give you the ability to actually use the blob detection tool and key out the green screen
Theoretically, If I were to get an external webcam and run an extension cord from my laptop, could this work for a small stage performance?
Yes ! you might need an active usb extension depending on the length of your cable though
@@BenjaminCarrierDesign awesome
Super helpful, thank you!
hey! very cool tutorial, thank you so much! Just a little off topic question? How do you make the little green circle appear to change the values? :)))
The middle mouse button, there is also a shortcut if you are on trackpad that can be set in the preferences :)
@@BenjaminCarrierDesign thank you :)
i have a macbook pro m1, is there anyway to do this on mac? using a different device (not kinect) for tracking.
Your webcam might work on m1, it works on my mbp
what a clever trick!
thanks a lot
For the Kinect version, was it a Xbox one or a 360 Kinect?
You could use both if you want!
it seems that nvidia background works super well too
hi :) thank u so much. if i wanna put new material to fill my the women in my video, how do i do that? or how i can put new background like u did in the begging of your example?
Either by using an over top to place it over the background or a multiply top to fill the shape of the person
How can I put the background?
with an over TOP you can place the image over another one, black background for example
Huuuu, is the trick to have a striped Tshirt :')? Like, it gives you more edges. I struggle with filling my chest with white.
humm good point I haven't tested it with white T-Shirt 😅 You might get better results by caching more images at the beginning and / or playing with the strength of the edge TOP
Pro trick: watch series on Flixzone. I've been using them for watching a lot of movies lately.
@Conor Leroy definitely, I have been watching on flixzone} for months myself =)
I really like that approach, but unfortunately it only works with white skin and bright colored or white clothes.
However you could use Math TOP's for limiting and normalizing instead of Level TOP's, and also adding some feedback at the end of the chain is improving this a little :)
Thank you for the feedback! Yes it seems like it would definitely give better results that way!
can i change the white color to red?
yes using a lookup OP you could change it to any color!
@@BenjaminCarrierDesign thank you