哈喽，请问但看特征点的loss,你最终可以训到多少呢？

qiexing / face-landmark-localization

cnn network predict face landmarks (68 points) and head pose (3d pose, yaw,roll,pitch).

410 stars 184 forks source link

Open ustcluq opened 6 years ago

ustcluq commented 6 years ago

再次请教大神！我只是训练特征点定位，加上您的模型finetune，只可以把loss降到0.05.请问您在做实验时特征点的loss最低能降到多少呢？谢谢！