MagicSource
MagicSource
how's the model performance if train on wholebody dataset?
Hi, pixie looks really impressive, just wonder, does there anyway to get a expression basis blend shapes in 51 dimensional same as apples' ARKit? this could be useful we want...
Will training process opensource for BlazeFace?
When using this snippet export yolov3 asff model out: ``` # Initiate model if args.asff: if backbone == 'mobile': from models.yolov3_mobilev2 import YOLOv3 print( "For mobilenet, we currently don't support...
 This is 500x500 input, when only 1 face, it runs 30fps+ on GTX1070, but speed drop fast when faces more than 15 and down to only 4 fps..,... How...
The model is tiny but speed is still not good though. It runs 0.4s per image, with some sample image from wider face. Is that normal?
how to run with audio driven?