Really excited for 0.7! Just wondering — will it improve the accuracy of foot recognition? And will this update include any option for users to train/customize the model themselves? That would be amazing!
