Partial calibration target detection for improved vehicle sensor calibration
Systems and methods provide for adaptive real-time streaming of Autonomous Vehicle (AV) data. In some embodiments, the AV can receive a request from the remote computing system for real-time streaming of a first type of AV data and adaptively streaming a second type of AV data when one or more streaming conditions are satisfied. The first type of AV data and the second type of AV data can be captured as raw data by sensors, actuators, transducers, and other components of the AV. The AV can stream the first type of AV data to the remote computing system in real-time for a first time period. When the AV determines the streaming conditions are satisfied, the AV can automatically determine the second type of AV data to stream to the remote computing system in real-time for a second time period.
1. A computer-implemented method comprising:
receiving a captured image from an image sensor coupled to a motor vehicle, the captured image including at least one representation of a partial calibration target;
identifying a plurality of quadrilaterals within the captured image;
determining a group of quadrilaterals corresponding to the at least one representation of the partial calibration target based on a relative proximity of the plurality of quadrilaterals to one another;
determining a number of rows and a number of columns of the group of quadrilaterals, wherein the number of rows and the number of columns of the group of quadrilaterals is dependent upon an applied image-based clustering technique;
determining corners of the at least one representation of the partial calibration target in the captured image based on the number of rows and the number of columns of the group of quadrilaterals; and
calibrating the image sensor based on the corners, the number of rows, and the number of columns of the at least one representation of the partial calibration target in the captured image.
2. The computer-implemented method of claim 1 , further comprising:
generating a plurality of binary images of the captured image by dilating the captured image based on a plurality of threshold values; and
determining a respective number of rows and a respective number of columns of the group of quadrilaterals corresponding to the at least one representation of the partial calibration target in the plurality of binary images,
wherein the number of rows and the number of columns is determined based on a combination of the respective number of rows and the respective number of columns of the group of quadrilaterals corresponding to the at least one representation of the partial calibration target in the plurality of binary images.
3. The computer-implemented method of claim 1 , further comprising:
determining neighboring quadrilaterals for each quadrilateral within the group of quadrilaterals; and
determining a row and a column of each quadrilateral within the group of quadrilaterals based on the neighboring quadrilaterals for each quadrilateral within the group of quadrilaterals,
wherein the number of rows and the number of columns is determined based on the row and the column of each quadrilateral within the group of quadrilaterals.
4. The computer-implemented method of claim 1 ,
wherein the at least one representation of the partial calibration target is located at an edge of the captured image.
5. A system, comprising:
one or more processors;
memory including instructions that, when executed by the one or more processors, cause the system to:
receive a captured image from an image sensor coupled to a motor vehicle, the captured image including at least one representation of a partial calibration target;
identify a plurality of quadrilaterals within the captured image;
determine a group of quadrilaterals corresponding to the at least one representation of the partial calibration target based on a relative proximity of the plurality of quadrilaterals to one another;
determine a number of rows and a number of columns of the group of quadrilaterals wherein the number of rows and the number of columns of the group of quadrilaterals is dependent upon an applied image-based clustering technique;
determine corners of the at least one representation of the partial calibration target in the captured image based on the number of rows and the number of columns of the group of quadrilaterals; and
calibrate the image sensor based on the corners, the number of rows, and the number of columns of the at least one representation of the partial calibration target in the captured image.
6. The system of claim 5 , further comprising further instructions that, when executed by the one or more processors, further cause the system to:
generate a plurality of binary images of the captured image by dilating the captured image based on a plurality of threshold values; and
determine a respective number of rows and a respective number of columns of the group of quadrilaterals corresponding to the at least one representation of the partial calibration target in the plurality of binary images,
wherein the number of rows and the number of columns is determined based on a combination of the respective number of rows and the respective number of columns of the group of quadrilaterals corresponding to the at least one representation of the partial calibration target in the plurality of binary images.
7. The system of claim 5 , further comprising further instructions that, when executed by the one or more processors, further cause the system to:
determine neighboring quadrilaterals for each quadrilateral within the group of quadrilaterals; and
determine a row and a column of each quadrilateral within the group of quadrilaterals based on the neighboring quadrilaterals for each quadrilateral within the group of quadrilaterals,
wherein the number of rows and the number of columns is determined based on the row and the column of each quadrilateral within the group of quadrilaterals.
8. The system of claim 5 , wherein the at least one representation of the partial calibration target is located at an edge of the captured image.
9. A non-transitory computer-readable storage medium including instructions that, upon being executed by one or more processors of a computing system, cause the computing system to:
receive a captured image from an image sensor coupled to a motor vehicle, the captured image including at least one representation of a partial calibration target;
identify a plurality of quadrilaterals within the captured image;
determine a group of quadrilaterals corresponding to the at least one representation of the partial calibration target based on a relative proximity of the plurality of quadrilaterals to one another;
determine a number of rows and a number of columns of the group of quadrilaterals wherein the number of rows and the number of columns of the group of quadrilaterals is dependent upon an applied image-based clustering technique;
determine corners of the at least one representation of the partial calibration target in the captured image based on the number of rows and the number of columns of the group of quadrilaterals; and
calibrate the image sensor based on the corners, the number of rows, and the number of columns of the at least one representation of the partial calibration target in the captured image.
10. The non-transitory computer-readable storage medium of claim 9 , further comprising further instructions that, when executed by the one or more processors, further cause the system to:
determine the captured image includes a plurality of representations of calibration targets;
determine a second number of rows and a second number of columns of a second group of quadrilaterals corresponding to a representation of a second calibration target included in the captured image;
determining second corners of the representation of the second calibration target in the captured image based on the second number of rows and the second number of columns of the second group of quadrilaterals; and
calibrating the image sensor based on the second corners, the second number of rows, and the second number of columns of the representation of the second calibration target in the captured image.
11. The non-transitory computer-readable storage medium of claim 10 , wherein the second calibration target corresponds to at least one of a full calibration target or a second partial calibration target.
12. The non-transitory computer-readable storage medium of claim 9 , wherein the at least one representation of the partial calibration target is occluded by a representation of a second object included in the captured image.