You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The automatic face recognition was often a bit finicky. It would place the face in the modeling space as a rather small feature in a seemingly random part of the screen (mostly somewhere close to the top, or close the bottom).
We need to crop the video stream captured from the webcam to match the shape of the LEM raster grid. However, it is difficult to see where to position the face in front of the camera for best results.
What would help is display the cropped though unprocessed camera stream (maybe with a rectangle showing the detected face) in another widget next to the simulated topography. One issue is that apparently it is not currently possible to resize ipywebrtc's camera widget.
The text was updated successfully, but these errors were encountered:
From Luca's feedback:
We need to crop the video stream captured from the webcam to match the shape of the LEM raster grid. However, it is difficult to see where to position the face in front of the camera for best results.
What would help is display the cropped though unprocessed camera stream (maybe with a rectangle showing the detected face) in another widget next to the simulated topography. One issue is that apparently it is not currently possible to resize ipywebrtc's camera widget.
The text was updated successfully, but these errors were encountered: