Understanding Inexplicit Utterances for Helper Robots Using Vision