Confidence determination of 3D point cloud data acquired by LIDAR sensor
Provided is an information processing device capable of determining accuracy of data acquired using an LIDAR sensor. An information processing device ( 10 ) according to the present disclosure includes: an image-capturing sensor ( 11 ); a learning unit ( 12 ) configured to learn, as learning data, learning image-capturing data and likelihood of a distance between points included in 3D point cloud data with correct answer in a region substantially the same region included in the learning image-capturing data, and to generate a learned model; and an estimation unit ( 13 ) configured to use the learned model to generate estimation data including likelihood of a distance between points included in estimation 3D point cloud data determined based on estimation image-capturing data, from the estimation image-capturing data acquired by the image-capturing sensor ( 11 ).
1. An information processing device comprising:
an image-capturing sensor;
at least one memory storing instructions; and
at least one processor configured to execute the instructions to:
determine learning image-capturing data of 3D point cloud data and a correct answer of a distance of points included in the 3D point cloud data;
generate a model, based on the learning image-capturing data, that estimates a likelihood of a distance of the points included in 3D point cloud data for a region in input image-capturing data that is substantially the same as a region included in the learning image-capturing data, the likelihood of the distance of the points representing an accuracy of the distance of the points in the 3D point cloud data;
generate estimation data including the likelihood of a the distance of the points included in the 3D point cloud data that is determined based on estimation image-capturing data, by inputting the estimation image-capturing data acquired by the image-capturing sensor to the learned model; and
providing the likelihood of the distance of the points included in the estimation data to each of the points in the 3D point cloud data as measured and acquired by a LIDAR sensor.
2. The information processing device according to claim 1 , wherein the at least one processor is further configured to execute the instructions to manage a correspondence relation between the points included in the 3D point cloud data with the correct answer and positions corresponding to pixels of the learning image-capturing data.
3. The information processing device according to claim 1 , wherein the likelihood of the distance of the points included in the 3D point cloud data is determined according to a result of comparison with a distance of surrounding points.
4. The information processing device according to claim 1 , wherein the 3D point cloud data as measured and acquired by the LIDAR sensor includes substantially the same region as a region in the estimation image-capturing data.
5. The information processing device according to claim 1 , wherein the at least one processor is further configured to execute the instructions to
generate likelihood estimation image data as the estimation data from labeled estimation data obtained by performing image processing on the estimation image-capturing data.
6. The information processing device according to claim 5 , wherein the at least one processor is further configured to execute the instructions to execute semantic segmentation as the image processing.
7. The information processing device according to claim 1 , wherein the at least one processor is further configured to execute the instructions to correct the likelihood of the distance of the points generated from the estimation image-capturing data according to at least one of an incident angle of laser beams irradiated from the LIDAR sensor with respect to a ground surface and a height of the LIDAR sensor from the ground surface.
8. A data generation method performed by a computer and comprising:
determining learning image-capturing data of 3D point cloud data and a correct answer of a distance of points included in the 3D point cloud data;
generating a model, based on the learning image-capturing data, that estimates a likelihood of a distance of the points included in 3D point cloud data for a region in input image-capturing data that is substantially the same as a region included in the learning image-capturing data, the likelihood of the distance of the points representing an accuracy of the distance of the points in the 3D point cloud data;
generating estimation data including the likelihood of a the distance of the points included in the 3D point cloud data that is determined based on estimation image-capturing data, by inputting the estimation image-capturing data acquired by the image-capturing sensor to the learned model; and
providing the likelihood of the distance of the points included in the estimation data to each of the points in the 3D point cloud data as measured and acquired by a LIDAR sensor.
9. A non-transitory computer-readable medium storing a program executable by a computer to perform processing comprising:
determining learning image-capturing data of 3D point cloud data and a correct answer of a distance of points included in the 3D point cloud data;
generating a model, based on the learning image-capturing data, that estimates a likelihood of a distance of the points included in 3D point cloud data for a region in input image-capturing data that is substantially the same as a region included in the learning image-capturing data, the likelihood of the distance of the points representing an accuracy of the distance of the points in the 3D point cloud data;
generating estimation data including the likelihood of a the distance of the points included in the 3D point cloud data that is determined based on estimation image-capturing data, by inputting the estimation image-capturing data acquired by the image-capturing sensor to the learned model; and
providing the likelihood of the distance of the points included in the estimation data to each of the points in the 3D point cloud data as measured and acquired by a LIDAR sensor.