Dynamic obstacle point cloud annotating method and apparatus, device and readable medium
A method comprises: collecting first point cloud data under a static scenario around a target collecting point; building a static background mesh model under the static scenario around the target collecting point according to the first point cloud data; collecting a second point cloud data of a target frame under a dynamic scenario after a dynamic obstacle moves around the target collecting point; annotating the point cloud of dynamic obstacle in the second point cloud data corresponding to the target frame, according to the static background mesh model. Through the technical solution of the present disclosure, it is feasible to automatically annotate the point cloud of the dynamic obstacle, effectively save manpower and annotation time spent in annotating the dynamic obstacle point cloud, and thereby effectively improve the efficiency of annotating the dynamic obstacle.
1. A dynamic obstacle point cloud annotating method, wherein the method comprises:
collecting first point cloud data under a static scenario around a target collecting point;
building a static background mesh model under the static scenario around the target collecting point according to the first point cloud data;
collecting a second point cloud data of a target frame under a dynamic scenario after a dynamic obstacle moves around the target collecting point;
annotating the point cloud of dynamic obstacle in the second point cloud data corresponding to the target frame, according to the static background mesh model,
wherein the building a static background mesh model under the static scenario around the target collecting point according to the first point cloud data specifically comprises:
performing mesh processing for the first point cloud data in a 3-dimensinal space to obtain a 3-dimensional mesh model;
setting a background attribute for the corresponding mesh according to whether or not each mesh in the 3-dimensional mesh model includes valid points in the first point cloud data, to obtain the static background mesh model under the static scenario around the target collecting point,
wherein the step of, according to whether each mesh in the 3-dimensional mesh model includes valid points in the first point cloud data, setting a background attribute for the corresponding mesh specifically comprises:
judging whether meshes in the 3-dimensional mesh model include valid points in the first point cloud data, and if yes, setting the background attribute of the corresponding mesh as 1; otherwise, setting the background attribute of the corresponding mesh as 0,
wherein the annotating the point cloud of dynamic obstacle in the second point cloud data corresponding to the target frame, according to the static background mesh model specifically comprises:
projecting each target point of the second point cloud data into the static background mesh model;
obtaining the background attribute of the mesh in the static background mesh model into which each target point is projected;
according to the background attribute of the mesh in the static background mesh model into which each target point is projected, recognizing whether the corresponding target point belongs to a point of the dynamic obstacle;
performing clustering processing for all points belonging to dynamic obstacles in the second point cloud data, to obtain the corresponding point cloud of dynamic obstacle.
2. The method according to claim 1 , wherein the step of, according to the background attribute of the mesh in the static background mesh model into which each target point is projected, recognizing whether the corresponding target point belongs to a point of the dynamic obstacle specifically comprises:
judging whether the background attribute of the mesh in the static background mesh model into which each target point is projected is 1 or 0;
if the background attribute of the mesh in the static background mesh model into which the corresponding target point is projected is 1, determining the target point as the background point of the static attribute;
if the background attribute of the mesh in the static background mesh model into which the corresponding target point is projected is 0, determining the target point as the point of the dynamic obstacle.
3. A computer device, wherein the device comprises:
one or more processors;
a storage for storing one or more programs;
a laser radar used to scan point cloud data under a static scenario around a target collecting point or under a dynamic scenario after a dynamic obstacle moves;
the one or more programs, when executed by said one or more processors, enable said one or more processors to implement a dynamic obstacle point cloud annotating method, wherein the method comprises:
collecting first point cloud data under a static scenario around a target collecting point;
building a static background mesh model under the static scenario around the target collecting point according to the first point cloud data;
collecting a second point cloud data of a target frame under a dynamic scenario after a dynamic obstacle moves around the target collecting point;
annotating the point cloud of dynamic obstacle in the second point cloud data corresponding to the target frame, according to the static background mesh model,
wherein the building a static background mesh model under the static scenario around the target collecting point according to the first point cloud data specifically comprises:
performing mesh processing for the first point cloud data in a 3-dimensinal space to obtain a 3-dimensional mesh model;
setting a background attribute for the corresponding mesh according to whether or not each mesh in the 3-dimensional mesh model includes valid points in the first point cloud data, to obtain the static background mesh model under the static scenario around the target collecting point,
wherein the step of, according to whether each mesh in the 3-dimensional mesh model includes valid points in the first point cloud data, setting a background attribute for the corresponding mesh specifically comprises:
judging whether meshes in the 3-dimensional mesh model include valid points in the first point cloud data, and if yes, setting the background attribute of the corresponding mesh as 1; otherwise, setting the background attribute of the corresponding mesh as 0,
wherein the annotating the point cloud of dynamic obstacle in the second point cloud data corresponding to the target frame, according to the static background mesh model specifically comprises:
projecting each target point of the second point cloud data into the static background mesh model;
obtaining the background attribute of the mesh in the static background mesh model into which each target point is projected;
according to the background attribute of the mesh in the static background mesh model into which each target point is projected, recognizing whether the corresponding target point belongs to a point of the dynamic obstacle;
performing clustering processing for all points belonging to dynamic obstacles in the second point cloud data, to obtain the corresponding point cloud of dynamic obstacle.
4. The computer device according to claim 3 , wherein the step of, according to the background attribute of the mesh in the static background mesh model into which each target point is projected, recognizing whether the corresponding target point belongs to a point of the dynamic obstacle specifically comprises:
judging whether the background attribute of the mesh in the static background mesh model into which each target point is projected is 1 or 0;
if the background attribute of the mesh in the static background mesh model into which the corresponding target point is projected is 1, determining the target point as the background point of the static attribute;
if the background attribute of the mesh in the static background mesh model into which the corresponding target point is projected is 0, determining the target point as the point of the dynamic obstacle.
5. A non-transitory computer readable medium on which a computer program is stored, wherein the program, when executed by a processor, implements a dynamic obstacle point cloud annotating method, wherein the method comprises:
collecting first point cloud data under a static scenario around a target collecting point;
building a static background mesh model under the static scenario around the target collecting point according to the first point cloud data;
collecting a second point cloud data of a target frame under a dynamic scenario after a dynamic obstacle moves around the target collecting point;
annotating the point cloud of dynamic obstacle in the second point cloud data corresponding to the target frame, according to the static background mesh model,
wherein the building a static background mesh model under the static scenario around the target collecting point according to the first point cloud data specifically comprises:
performing mesh processing for the first point cloud data in a 3-dimensinal space to obtain a 3-dimensional mesh model;
setting a background attribute for the corresponding mesh according to whether or not each mesh in the 3-dimensional mesh model includes valid points in the first point cloud data, to obtain the static background mesh model under the static scenario around the target collecting point,
wherein the step of, according to whether each mesh in the 3-dimensional mesh model includes valid points in the first point cloud data, setting a background attribute for the corresponding mesh specifically comprises:
judging whether meshes in the 3-dimensional mesh model include valid points in the first point cloud data, and if yes, setting the background attribute of the corresponding mesh as 1; otherwise, setting the background attribute of the corresponding mesh as 0,
wherein the annotating the point cloud of dynamic obstacle in the second point cloud data corresponding to the target frame, according to the static background mesh model specifically comprises:
projecting each target point of the second point cloud data into the static background mesh model;
obtaining the background attribute of the mesh in the static background mesh model into which each target point is projected;
according to the background attribute of the mesh in the static background mesh model into which each target point is projected, recognizing whether the corresponding target point belongs to a point of the dynamic obstacle;
performing clustering processing for all points belonging to dynamic obstacles in the second point cloud data, to obtain the corresponding point cloud of dynamic obstacle.
6. The non-transitory computer readable medium according to claim 5 , wherein the step of, according to the background attribute of the mesh in the static background mesh model into which each target point is projected, recognizing whether the corresponding target point belongs to a point of the dynamic obstacle specifically comprises:
judging whether the background attribute of the mesh in the static background mesh model into which each target point is projected is 1 or 0;
if the background attribute of the mesh in the static background mesh model into which the corresponding target point is projected is 1, determining the target point as the background point of the static attribute;
if the background attribute of the mesh in the static background mesh model into which the corresponding target point is projected is 0, determining the target point as the point of the dynamic obstacle.