HartyBart opened this issue on Nov 15, 2023 ยท 2 posts
HartyBart posted Wed, 15 November 2023 at 1:49 AM
Face Landmark, new, free and open source. It looks at a webcam video of your face, automatically extracts and cleans up the 'blend shapes' needed for facial animation in 3D, then drops these to a simple .CSV text file (suitable for LiveLink, in this case).
Open source, and I'd suggest it's a prime candidate for adaptation to do simple "your webcam to Poser" facial motion capture, which would greatly enhance the appeal of Poser. In fact, since the source Face_Landmarker.py and face_geometry.py Python files are available, one might theoretically rework it to output 'straight to Poser format'?
https://github.com/Qaanaaq/Face_Landmark_Link/
Learn the Secrets of Poser 11 and Line-art Filters.
an0malaus posted Thu, 16 November 2023 at 8:12 PM
@HartyBart thanks for the heads-up! This might indeed be applicable to Poser!
I suggested to the author that the missing tongue positions could possibly be extracted using Python Speech Recognition Library and a phoneme dictionary, with the phonemes giving a clue to tongue positions.
Very interesting!
Verbosity: Profusely promulgating Graham's number epics of complete and utter verbiage by the metric monkey barrel.