open world object detection and discoveryCreated: 21-07-2022 Forks: 2 Watchers: 0 Stars: 0
Object detection is a well-known computer vision technique for locating instances of objects in images. Open World Object Detection refers to tackling object detection tasks in an open-world setting where the task is to detect a known set of object categories while simultaneously identifying unknown objects. Moreover, it is required to identify instances of unknown objects and subsequently learn to recognize them when training data progressively arrive. An oracle (eg: - a human annotator) is needed to introduce new class labels for the identified unknowns to classify those to relevant object classes in the next increments of learning steps.