Image-based parking recognition and navigation

A computer-implemented method comprising:

• capturing, using one or more cameras at a parking lot, first image data associated with the vehicle;
• capturing, using the one or more cameras at the parking lot, second image data associated with the parking lot;
• providing the image data associated with the parking lot to a computer vision machine learning model to determine a condition of the parking lot indicating at least dimensions of one or more available parking spaces;
• determining the condition of the parking lot based on a first output of the computer vision machine learning model;
• determining if the parking lot has the available parking space to fit the vehicle using a second computer vision machine learning model, the second computer vision machine learning model running on a processor located proximate to an entrance of the parking lot, an input to the second computer vision learning model including the user preferences and dimensions of the vehicle determined based on the first image data;
• and
• in response to determining the parking lot has the available parking space to fit the vehicle, transmitting an instruction to an access control mechanism at the parking lot to permit access to the available parking space by the vehicle.

A non-transitory computer-readable storage medium comprising instructions stored therein, which when executed by one or more processors, cause the processors to perform operations comprising:

• capturing, using one or more cameras at a parking lot, first image data associated with the vehicle;
• capturing, using the one or more cameras at the parking lot, second image data associated with the parking lot;
• providing the image data associated with the parking lot to a computer vision machine learning model to determine a condition of the parking lot indicating at least dimensions of one or more available parking spaces;
• determining the condition of the parking lot based on a first output;
• determining if the parking lot has the available parking space to fit the vehicle using a second computer vision machine learning model, the second computer vision machine learning model running on a processor located proximate to an entrance of the parking lot, an input to the second computer vision learning model including the user preferences and dimensions of the vehicle determined based on the first image data;
• and
• in response to determining the parking lot has the available parking space to fit the vehicle, transmitting an instruction to an access control mechanism at the parking lot to permit access to the available parking space by the vehicle.