Open patrikhuber opened 9 years ago
I added a model to master
that's trained on the 29 COFW landmarks and achieves a state-of-the-art error of around 0.07. The COFW-29 set is better to evaluate as it's a standardised set, not like the me17 on LFPW, whose images are largely not available anymore. Still I should do a test on LFPW at some point.
I'd like to support the popular me17 landmark set that's used in a lot of papers (e.g. Zhou et al. and ours). However, some landmarks are different from the ibug-68 set and above papers also use the original LFPW set (or a subset of it), and not the ibug release of LFPW.
We should thus support training in the original LFPW as well.