Parking spot detection

A computerized apparatus having a hardware processor, the processor the hardware processor being coupled to a memory, being hardware processor adapted to perform the steps of:

• obtaining an elevated perception map of a surrounding area around a vehicle, each pixel in the elevated perception map is associated with a predetermined relative location to the vehicle, wherein the elevated perception map comprises a plurality of functional layers, wherein values of pixels at different layers indicate an infrastructure segment or object located at corresponding relative locations to the vehicle;
• performing parking slot object detection in the elevated perception map, wherein the parking slot object detection is performed using an Artificial Neural Network (ANN), whereby obtaining one or more detected parking slot objects, wherein the plurality of functional layers of the elevated perception map comprises a vehicle object layer that is indicative of other vehicles that are present in the surrounding area of the vehicle, whereby the parking slot object detection is configured to classify parking slots as occupied or vacant based on the other vehicles, wherein the vehicle object layer is determined based on real-time sensor readings at the surrounding area of the vehicle, wherein the vehicle object layer is determined based on a filtration of vehicles that travel at a speed above a predetermined threshold speed;
• and
• outputting the one or more detected parking slot objects.

A computer program product comprising a non-transitory computer readable storage medium retaining program instructions, which program instructions when read by a processor, cause the processor to perform a method comprising:

• obtaining an elevated perception map of a surrounding area around a vehicle, each pixel in the elevated perception map is associated with a predetermined relative location to the vehicle, wherein the elevated perception map comprises a plurality of functional layers, wherein values of pixels at different layers indicate an infrastructure segment or object located at corresponding relative locations to the vehicle;
• performing parking slot object detection in the elevated perception map, wherein the parking slot object detection is performed using an Artificial Neural Network (ANN), whereby obtaining one or more detected parking slot objects, wherein the plurality of functional layers of the elevated perception map comprises a vehicle object layer that is indicative of other vehicles that are present in the surrounding area of the vehicle, whereby the parking slot object detection is configured to classify parking slots as occupied or vacant based on the other vehicles, wherein the vehicle object layer is determined based on real-time sensor readings at the surrounding area of the vehicle, wherein the vehicle object layer is determined based on a filtration of vehicles that travel at a speed above a predetermined threshold speed;
• and
• outputting the one or more detected parking slot objects.