Open wshilton opened 1 year ago
For FAE, candidate starting point is Google's MediaPipe face landmark detection.
Originally posted by @wshilton in https://github.com/wshilton/andrew/issues/4#issuecomment-1546723928
Initial implementation of pre-trained face landmark detection network works well. These facilities are the groundwork for facial attribute estimation.
Originally posted by @wshilton in https://github.com/wshilton/andrew/issues/4#issuecomment-1546723928
Initial implementation of pre-trained face landmark detection network works well. These facilities are the groundwork for facial attribute estimation.