Systems and methods for detecting carried objects to adapt vehicle access
System, methods, and other embodiments described herein relate to adapting vehicle access by detecting a person carrying an object. In one embodiment, a method includes detecting a person near a vehicle for gaining access. The method also includes scanning the person for an object using a radar of the vehicle, wherein information from the radar indicates densities of the person and the object. Upon detecting the object using the densities, the method also includes adapting the access to a compartment of the vehicle.
1. A detection system for controlling access to a vehicle, comprising:
a processor; and
a memory storing instructions that, when executed by the processor, cause the processor to:
detect a person near the vehicle to gain the access;
scan the person for an object using a radar of the vehicle, wherein processing information from the radar indicates densities of the person and the object and voxels about the object; and
upon detection of the object using the densities, unlock a compartment of the vehicle that can fit the object according to the voxels and depth points about the object that are grouped.
2. The detection system of claim 1 , wherein the instructions to scan the person further include instructions to estimate, using a machine learning (ML) model, the depth points according to the information and detection of the object as an occlusion in a foreground associated with the person using the depth and the densities.
3. The detection system of claim 1 , further including instructions to:
compute, using a machine learning (ML) model, the voxels representing a volume of the object according to one of millimeter and ultra-wideband waves returned from the radar;
track movement and intensities of the voxels associated with the object; and
open the compartment according to the volume and patterns of the movement associated with the person.
4. The detection system of claim 3 , further including instructions to:
determine a label for the object according to the volume and the densities, wherein the label indicates a type of the object; and
open a door on a condition that the label indicates the object as cargo.
5. The detection system of claim 1 , further including instructions to:
scan, by the radar subsequent to the unlock, the compartment for the object; and
upon the person leaving the vehicle, initiate an alarm as a notification to take the object.
6. The detection system of claim 1 , wherein the unlock includes one of: opening a trunk, opening a door, unlocking the trunk, and unlocking the door of the vehicle.
7. The detection system of claim 1 , wherein the instructions to detect the person further include instructions to authorize the unlock using one of: identification data from a mobile device, a digital key, and a key fob.
8. The detection system of claim 1 , wherein the densities indicate one of human tissue, bone, and liquid materials.
9. A non-transitory computer-readable medium for controlling access to a vehicle comprising:
instructions that when executed by a processor cause the processor to:
detect a person near the vehicle to gain the access;
scan the person for an object using a radar of the vehicle, wherein processing information from the radar indicates densities of the person and the object and voxels about the object; and
upon detection of the object using the densities, unlock a compartment of the vehicle that can fit the object according to the voxels and depth points about the object that are grouped.
10. The non-transitory computer-readable medium of claim 9 , wherein the instructions to scan the person further include instructions to estimate, using a machine learning (ML) model, the depth points according to the information and detection of the object as an occlusion in a foreground associated with the person using the depth and the densities.
11. The non-transitory computer-readable medium of claim 9 , further including instructions to:
compute, using a machine learning (ML) model, the voxels representing a volume of the object according to one of millimeter and ultra-wideband waves returned from the radar;
track movement and intensities of the voxels associated with the object; and
open the compartment according to the volume and patterns of the movement associated with the person.
12. The non-transitory computer-readable medium of claim 11 , further including instructions to:
determine a label for the object according to the volume and the densities, wherein the label indicates a type of the object; and
open a door on a condition that the label indicates the object as cargo.
13. A method comprising:
detecting a person near a vehicle for gaining access;
scanning the person for an object using a radar of the vehicle, wherein processing information from the radar indicates densities of the person and the object and voxels about the object; and
upon detecting the object using the densities, unlocking a compartment of the vehicle that can fit the object according to the voxels and depth points about the object that are grouped.
14. The method of claim 13 , wherein scanning the person further includes estimating, using a machine learning (ML) model, the depth points according to the information and detecting the object as an occlusion in a foreground associated with the person using the depth and the densities.
15. The method of claim 13 , further comprising:
computing, using a machine learning (ML) model, the voxels representing a volume of the object according to one of millimeter and ultra-wideband waves returned from the radar;
tracking movement and intensities of the voxels associated with the object; and
opening the compartment according to the volume and patterns of the movement associated with the person.
16. The method of claim 15 , further comprising:
determining a label for the object according to the volume and the densities, wherein the label indicates a type of the object; and
opening a door on a condition that the label indicates the object as cargo.
17. The method of claim 13 , further comprising:
scanning, by the radar subsequent to the unlock, the compartment for the object; and
upon the person leaving the vehicle, initiating an alarm as a notification to take the object.
18. The method of claim 13 , wherein the unlock includes one of:
opening a trunk, opening a door, unlocking the trunk, and unlocking the door of the vehicle.
19. The method of claim 13 , wherein detecting the person further includes authorizing unlocking of the compartment using one of: identification data from a mobile device, a digital key, and a key fob.
20. The method of claim 13 , wherein the densities indicate one of human tissue, bone, and liquid materials.