Models for stop sign database creation
An improved system and method of determining presence of stop signs at intersection based on multiple types of data including ground truth data, telemetry data, and/or trip stills gathered from vehicle devices. The data may be analyzed and aggregated by one or more models and/or neural networks to develop a prediction regarding presence of stop signs at individual intersections.
1. A method performed by a computing system having one or more hardware computer processors and one or more non-transitory computer readable storage device storing software instructions executable by the computing system, the method comprising:
accessing map data including indications of points of interest within a geographic region;
accessing a plurality of trip stills data points each indicating at least a trip stills image obtained from a vehicle camera and an associated point of interest in the map data;
accessing a plurality of telemetry data points each indicating at least a speed of a vehicle and an associated point of interest in the map data; and
for individual points of interest:
evaluating a neural network based on the telemetry data points associated with the individual point of interest to determine a telemetry probability for the individual point of interest;
evaluating an object detection model for one or more trip stills data points associated with the individual point of interest to determine a corresponding one or more trip stills probabilities indicating likelihood of an object of interest in the trip stills image of the trip stills data point; and
determining a likelihood of the object of interest being present at the individual point of interest based on at least the telemetry probability for the individual point of interest and the one or more trip stills probabilities associated with the point of interest.
2. The method of claim 1 , wherein if both a telemetry probability and a predetermined quantity of trip stills probabilities are available for an individual point of interest, the likelihood of the object of interest being present at the individual point of interest is based on a combined prediction representative of aggregation of the telemetry probability and the trip stills probabilities.
3. The method of claim 2 , wherein the combined prediction indicates presence of an object of interest if the telemetry probability is above a first threshold and less than a minimum quantity of trip stills probabilities for the point of interest are available.
4. The method of claim 3 , wherein the first threshold is 95%.
5. The method of claim 3 , wherein the likelihood of the object of interest being present at the individual point of interest indicates presence of an object of interest if at least one trip stills probability is above a second threshold.
6. The method of claim 5 , wherein the second threshold is 90%.
7. The method of claim 6 , wherein the likelihood of the object of interest being present at the individual point of interest indicates presence of an object of interest if at least two trip stills probabilities are above a third threshold.
8. The method of claim 7 , wherein the third threshold is 80%.
9. The method of claim 1 , further comprising, for each point of interest:
evaluating each of a plurality of trip stills probabilities associated with the individual point of interest; and
determining an overall trip stills probability for the individual point of interest based on the plurality of trip stills probabilities.
10. The method of claim 9 , wherein the overall trip stills probability indicates a quantity of trip stills probabilities associated with the individual point of interest.
11. The method of claim 1 , wherein a first individual point of interest indicates a first direction of travel associated with two intersecting streets.
12. The method of claim 11 , wherein a second individual point of interest indicates a second direction of travel associated with the two intersecting streets.
13. The method of claim 1 , wherein the objects of interest comprise stop signs.
14. The method of claim 1 , wherein the points of interest comprise intersections.
15. A computing system comprising:
a hardware computer processor;
a non-transitory computer readable medium having software instructions stored thereon, the software instructions executable by the hardware computer processor to cause the computing system to perform operations comprising:
accessing map data including indications of points of interest within a geographic region;
accessing a plurality of trip stills data points each indicating at least a trip stills image obtained from a vehicle camera and an associated point of interest in the map data;
accessing a plurality of telemetry data points each indicating at least a speed of a vehicle and an associated point of interest in the map data; and
for individual points of interest:
evaluating a neural network based on the telemetry data points associated with the individual point of interest to determine a telemetry probability for the individual point of interest;
evaluating an object detection model for one or more trip stills data points associated with the individual point of interest to determine a corresponding one or more trip stills probabilities indicating likelihood of an object of interest in the trip stills image of the trip stills data point; and
determining a likelihood of the object of interest being present at the individual point of interest based on at least the telemetry probability for the individual point of interest and the one or more trip stills probabilities associated with the point of interest.
16. The computing system of claim 15 , wherein if both a telemetry probability and a predetermined quantity of trip stills probabilities are available for an individual point of interest, the likelihood of the object of interest being present at the individual point of interest is based on a combined prediction representative of aggregation of the telemetry probability and the trip stills probabilities.
17. The computing system of claim 16 , wherein the combined prediction indicates presence of an object of interest if the telemetry probability is above a first threshold and less than a minimum quantity of trip stills probabilities for the point of interest are available.
18. The computing system of claim 17 , wherein the first threshold is 95%.
19. The computing system of claim 17 , wherein the likelihood of the object of interest being present at the individual point of interest indicates presence of an object of interest if at least one trip stills probability is above a second threshold.
20. The computing system of claim 19 , wherein the second threshold is 90%.