Example implementations relate to vehicle occupancy confirmation. An example implementation involves receiving, at a computing system from a camera positioned inside a vehicle, an image representing an occupancy within the vehicle. The implementation further involves, responsive to receiving the image, displaying the image on a display interface, and receiving an operator input confirming the occupancy meets a desired occupancy. The implementation additionally includes transmitting an occupancy confirmation from the computing system to the vehicle. In some instances, in response to receiving the occupancy confirmation, the vehicle executes an autonomous driving operation.