You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When a model should be trained on two datasets, which are similiar but labeled with different classes, we want to be able to align those classes. For this, we leverage Zero-Shot Image Classification models. Based on a one-to-many class mapping, the models can propose class changes for individual detections.
The text was updated successfully, but these errors were encountered:
When a model should be trained on two datasets, which are similiar but labeled with different classes, we want to be able to align those classes. For this, we leverage Zero-Shot Image Classification models. Based on a one-to-many class mapping, the models can propose class changes for individual detections.
The text was updated successfully, but these errors were encountered: