Object Detectors refer to models or algorithms designed to identify and localize objects within images or scenes. These detectors are trained using discriminative objective functions, aiming to distinguish between positive instances (objects of interest) and negative instances (background or irrelevant objects).


Generating Enhanced Negatives for Training Language-Based Object Detectors

The recent progress in language-based open-vocabulary object detection can be largely attributed to finding better ways of leveraging large-scale data with free-form text annotations. Training such models with a discriminative objective function has proven successful, but requires good positive and negative samples.