clmtrackr
clmtrackr copied to clipboard
My beard is confusing it :)
It thinks that the top of my beard is my upper lip. Is there anything i can do other than saving my beard?
Could you go this page, http://auduno.github.io/clmtrackr/clm_video_responses.html , and check if selecting a different response mode makes any difference? I'm especially curious if the choice 'Type : "single", List: ["lbp"]' makes any difference.
The single sobel setting found my upper lip but lost my lower lip. The 'Type : "single", List: ["lbp"]' didn't make any difference.
I think I have to retrain my model to see if that improves the tracking with beards. I'm a bit busy at the moment, but I'll see if I can do it sometime over the next weeks.
Sobel also worked better for me, but failed to track the mouth sometimes.
I wanted to ask if there is a "quick" work around where I could only make it track eyes and nose because for my project I would only need the upper parts.
This happens to me all the time. With my beard and static images I've used for testing.
@auduno thanks in advance for the retraining and all your work on this project.