Object detection and detection confidence suitable for autonomous driving

A system comprising:

• one or more processing units;
• one or more memory devices storing instructions thereon that, when executed using the one or more processing units, cause the one or more processing units to execute operations comprising:
• applying, to one or more machine learning models (MLMs), image data representative of one or more frames that depict a field of view of at least one sensor, wherein the one or more MLMs are trained to infer spatial element regions each correspond to the object based at least on the spatial element regions each falling at least partially in a region corresponding to the object in the field of view;
• receiving, from the one or more MLMs, first data representative of coverage values indicating likelihoods the spatial element regions correspond to the object and second data representative of predictions indicative of at least one dimension of the region;
• and
• determining, one or more dimensions of an object region of the object based at least on the likelihoods the spatial element regions correspond to the object and the predictions indicative of the at least one dimension of the region.