Systems and methods for 2D detections and tracking
According to some embodiments, a method includes accessing a video generated by a camera in a physical environment. The method further includes identifying, from a first video frame of the video, a first object of interest corresponding to a physical object in the physical environment. The method further includes storing a record for the first object of interest that includes a unique identifier. The method further includes identifying a second object of interest from a second video frame. The method further includes comparing variables of the first object of interest to variables of the second object of interest. The method further includes determining that the variables of the first object of interest match the variables of the second object of interest and then assigning the unique identifier to the second object of interest. The method further includes updating the record to include the second object of interest.
1. A system comprising:
one or more memory units; and
one or more computer processors communicatively coupled to the one or more memory units and configured to:
access a video generated by a camera located within a physical environment;
identify, by analyzing a first video frame of the video, a first object of interest in the first video frame, the first object of interest corresponding to a physical object that is physically located within the physical environment;
evaluate the identified first object of interest using a predetermined confidence threshold;
store, in the one or more memory units, a record for the first object of interest, the record comprising a unique identifier for the first object of interest;
identify, by analyzing a second video frame of the video, a second object of interest in the second video frame;
compare variables of the first object of interest to variables of the second object of interest;
determine, based on the comparison, that the variables of the first object of interest match the variables of the second object of interest and in response, assign the unique identifier for the first object of interest to the second object of interest;
update the record to include the second object of interest;
display, in a graphical user interface, a virtual three-dimensional (3D) environment that corresponds to the physical environment;
display a first virtual object in the virtual 3D environment that corresponds to the first object of interest; and
display a second virtual object in the virtual 3D environment that corresponds to the second object of interest, wherein the display of the first and second virtual objects within the virtual 3D environment indicates movement of the physical object within the physical environment.
2. The system of claim 1 , wherein identifying the first and second objects of interest comprises utilizing a convolution neural network architecture.
3. The system of claim 1 , wherein identifying the first and second objects of interest comprises identifying an object type for each of the first and second objects of interest.
4. The system of claim 3 , wherein the variables of the first object of interest are determined to match the variables of the second object of interest when:
the object type of the second object of interest matches the object type of the first object of interest;
the second object of interest is determined to be within a predetermined distance of the first object of interest; and
the velocity of the second object of interest is determined to be within a predetermined amount of the velocity of the first object of interest.
5. The system of claim 1 , wherein the variables of the first and second objects of interest comprise:
a proximity;
a velocity; and
an object type.
6. The system of claim 1 , wherein the first and second objects of interest are:
people;
automobiles; or
animals.
7. A method by a computing system, the method comprising:
accessing a video generated by a camera located within a physical environment;
identifying, by analyzing a first video frame of the video, a first object of interest in the first video frame, the first object of interest corresponding to a physical object that is physically located within the physical environment;
evaluating the identified first object of interest using a predetermined confidence threshold;
storing, in the one or more memory units, a record for the first object of interest, the record comprising a unique identifier for the first object of interest;
identifying, by analyzing a second video frame of the video, a second object of interest in the second video frame;
comparing variables of the first object of interest to variables of the second object of interest;
determining, based on the comparison, that the variables of the first object of interest match the variables of the second object of interest and in response, assigning the unique identifier for the first object of interest to the second object of interest;
updating the record to include the second object of interest;
displaying, in a graphical user interface, a virtual three-dimensional (3D) environment that corresponds to the physical environment;
displaying a first virtual object in the virtual 3D environment that corresponds to the first object of interest; and
displaying a second virtual object in the virtual 3D environment that corresponds to the second object of interest, wherein the display of the first and second virtual objects within the virtual 3D environment indicates movement of the physical object within the physical environment.
8. The method of claim 7 , wherein identifying the first and second objects of interest comprises utilizing a convolution neural network architecture.
9. The method of claim 7 , wherein identifying the first and second objects of interest comprises identifying an object type for each of the first and second objects of interest.
10. The method of claim 7 , wherein the variables of the first and second objects of interest comprise:
a proximity;
a velocity; and
an object type.
11. The method of claim 10 , wherein the variables of the first object of interest are determined to match the variables of the second object of interest when:
the object type of the second object of interest matches the object type of the first object of interest;
the second object of interest is determined to be within a predetermined distance of the first object of interest; and
the velocity of the second object of interest is determined to be within a predetermined amount of the velocity of the first object of interest.
12. The method of claim 7 , wherein the first and second objects of interest are:
people;
automobiles; or
animals.
13. One or more computer-readable non-transitory storage media embodying instructions that, when executed by a processor, cause the processor to perform operations comprising:
accessing a video generated by a camera located within a physical environment;
identifying, by analyzing a first video frame of the video, a first object of interest in the first video frame, the first object of interest corresponding to a physical object that is physically located within the physical environment;
evaluating the identified first object of interest using a predetermined confidence threshold;
storing, in the one or more memory units, a record for the first object of interest, the record comprising a unique identifier for the first object of interest;
identifying, by analyzing a second video frame of the video, a second object of interest in the second video frame;
comparing variables of the first object of interest to variables of the second object of interest;
determining, based on the comparison, that the variables of the first object of interest match the variables of the second object of interest and in response, assigning the unique identifier for the first object of interest to the second object of interest;
updating the record to include the second object of interest;
displaying, in a graphical user interface, a virtual three-dimensional (3D) environment that corresponds to the physical environment;
displaying a first virtual object in the virtual 3D environment that corresponds to the first object of interest; and
displaying a second virtual object in the virtual 3D environment that corresponds to the second object of interest, wherein the display of the first and second virtual objects within the virtual 3D environment indicates movement of the physical object within the physical environment.
14. The one or more computer-readable non-transitory storage media of claim 13 , wherein identifying the first and second objects of interest comprises utilizing a convolution neural network architecture.
15. The one or more computer-readable non-transitory storage claim 13 , wherein identifying the first and second objects of interest comprises identifying an object type for each of the first and second objects of interest.
16. The one or more computer-readable non-transitory storage claim 13 , wherein the variables of the first and second objects of interest comprise:
a proximity;
a velocity; and
an object type.
17. The one or more computer-readable non-transitory storage claim 16 , wherein the variables of the first object of interest are determined to match the variables of the second object of interest when:
the object type of the second object of interest matches the object type of the first object of interest;
the second object of interest is determined to be within a predetermined distance of the first object of interest; and
the velocity of the second object of interest is determined to be within a predetermined amount of the velocity of the first object of interest.