Try to use 2 models: one optimized for 25 f/s, another for 14 f/s

#17
No description provided.
xi0v changed pull request status to merged

Sign up or log in to comment