Perception System for Autonomous Vehicles

An example method includes generating, based on sensor data representing an environment, a plurality of proposed detection outputs corresponding to a plurality of positions in a representation of the environment, wherein a detection output of the plurality of proposed detection outputs indicates a proposed detected object in the environment and comprises an initial value corresponding to an initial likelihood for an attribute of the proposed detected object; generating, by a component that receives input including local context data and the initial value, an updated value for the attribute, wherein the updated value corresponds to an updated likelihood for the attribute, and wherein the local context data comprises, for a location in the environment associated with the proposed detected object, a portion of the sensor data or a portion of latent feature data; and generating an object detection output based on the updated value.