Abstract
A method includes capturing, by a camera in a robotic system, a first image of an object at a first location. The method also includes, identifying, by a processor in the robotic system, whether one or more features are present in the first image of the object using an object detection model. In response to determining that at least one feature of the one or more features are present in the first image of the object, the method further includes instructing, by the processor, a robotic agent to move based on a class of the at least one feature that is present in the first image. The method may also include capturing, by the camera, a second image of the object at a second location.