Automatic labeling of objects in sensor data
Aspects of the disclosure provide for automatically generating labels for sensor data. For instance first sensor data for a first vehicle is identified. The first sensor data is defined in both a global coordinate system and a local coordinate system for the first vehicle. A second vehicle is identified based on a second location of the second vehicle within a threshold distance of the first vehicle within the first timeframe. The second vehicle is associated with second sensor data that is further associated with a label identifying a location of an object, and the location of the object is defined in a local coordinate system of the second vehicle. A conversion from the local coordinate system of the second vehicle to the local coordinate system of the first vehicle may be determined and used to transfer the label from the second sensor data to the first sensor data.
1. A method comprising:
identifying, by one or more processors, first sensor data captured by one or more first sensors at a first location during a first timeframe, wherein the first sensor data is associated with a first label identifying a characteristic of an object and a location of the object defined in a first coordinate system of the first sensor data;
identifying, by the one or more processors, a vehicle at a second location that is within a threshold distance of the first location, the vehicle having one or more second sensors positioned thereon;
identifying, by the one or more processors, second sensor data captured by the one or more second sensors of the vehicle at the second location during a second timeframe that overlaps with the first timeframe, wherein the one or more second sensors of the vehicle are different than the one or more first sensors;
determining, by the one or more processors, a conversion from the first coordinate system of the first sensor data to a second coordinate system of the vehicle, wherein a first origin of the first coordinate system is different from a second origin of the second coordinate system; and
generating, by the one or more processors based on the conversion and the first sensor data, a second label associated with the second sensor data, wherein the second label includes the characteristic of the object.
2. The method of claim 1 , wherein the characteristic of the object is a type of the object.
3. The method of claim 1 , wherein the threshold distance is a radial distance from the second location.
4. The method of claim 1 , wherein the threshold distance corresponds to an area beyond a sensor range of at least one of the one or more second sensors of the vehicle.
5. The method of claim 1 , wherein the threshold distance includes an orientation component.
6. The method of claim 1 , wherein determining the conversion includes determining a difference between global coordinate system coordinates of the first origin of the first coordinate system and global coordinate system coordinates of the second origin of the second coordinate system.
7. The method of claim 6 , wherein applying the conversion includes subtracting the difference from the location of the object.
8. The method of claim 1 , wherein the first timeframe and the second timeframe are synced via GPS timing signals.
9. The method of claim 1 , wherein the one or more first sensors are positioned on a different vehicle other than the vehicle.
10. A system comprising one or more processors configured to:
identify first sensor data captured by one or more first sensors at a first location during a first timeframe, wherein the first sensor data is associated with a first label identifying a characteristic of an object and a location of the object defined in a first coordinate system of the first sensor data;
identify a vehicle at a second location that is within a threshold distance of the first location, the vehicle having one or more second sensors positioned thereon;
identify second sensor data captured by the one or more second sensors of the vehicle at the second location during a second timeframe that overlaps with the first timeframe, wherein the one or more second sensors of the vehicle are different than the one or more first sensors;
determine a conversion from the first coordinate system of the first sensor data to a second coordinate system of the vehicle, wherein a first origin of the first coordinate system is different from a second origin of the second coordinate system; and
generate a second label associated with the second sensor data based on the conversion and the first sensor data, wherein the second label includes the characteristic of the object.
11. The system of claim 10 , wherein the characteristic of the object is a type of the object.
12. The system of claim 10 , wherein the threshold distance is a radial distance from the second location.
13. The system of claim 10 , wherein the threshold distance corresponds to an area beyond a sensor range of at least one of the one or more second sensors of the vehicle.
14. The system of claim 10 , wherein the threshold distance includes an orientation component.
15. The system of claim 10 , wherein the one or more processors are further configured to determine the conversion by determining a difference between global coordinate system coordinates of the first origin of the first coordinate system and global coordinate system coordinates of the second origin of the second coordinate system.
16. The system of claim 15 , wherein applying the conversion includes subtracting the difference from the location of the object.
17. The system of claim 10 , wherein the first timeframe and the second timeframe are synced via GPS timing signals.
18. The system of claim 10 , wherein the one or more first sensors are positioned on a different vehicle other than the vehicle.