Firstly, conceptually, LIDAR is a laser technology that can capture objects and three-dimensional surfaces in space. After scanning, a data model can be obtained, however, the original scanning model generally has a high number of faces and its uv is not considered continuous, the scanning model I got is using each face is a separate uv block, this uv block corresponds to a piece of colour on the texture map, maybe this is more convenient for the automatic generation of uv and texture maps when scanning, and the automatic generation process may produce a lot of fragmented and scattered Each texture corresponds to a material ball. It is possible to generate a low model of the scanned model by surface reduction software, but if you don’t merge and re-process the uv and texture of the generated low model, it will lead to too many material spheres and textures, and there are too many uv seams and overlapping vertices, which makes it difficult to do normal baking.So we probably won’t be doing too much with the materials of the model, and will probably continue to use a point model rather than a face model at this stage.And the treatment of the point model can be summarised in three steps：
a. Importing point cloud data:
Firstly, you need to import the point cloud data scanned by the LiDAR into the computer software of your choice. Many 3D modelling software support importing point cloud data, such as MeshLab, CloudCompare, etc.
b. Cleaning and filtering:
Lidar scanned point cloud data may contain noisy or irrelevant points that need to be cleaned and filtered. This can be done by removing outlier points, performing smoothing operations, or applying other filters.
c. Point Cloud Reconstruction:
After cleaning and filtering, you can use point cloud reconstruction algorithms to convert the point cloud data into a surface mesh, which means that the points in the point cloud are connected to form a surface. This can be done with algorithms such as Poisson Reconstruction, Marching Cubes.
And all the above steps can be simply realised in the software accompanying the radar system. And after the above steps are completed we need to import the model into Maya or Blender to set up the track animation. The tentative initial approach now is to set up a first person view camera as the subject.The first thing to do is to create the path curve, which will animate the object that will move along the track, this can be done using the “EP Curve Tool” in the “Create” menu to draw the curve freely. Then it’s time to set up the path animation: select the camera and choose “Animate” -> “Motion Paths” -> “Attach to Motion Path” in the menu bar. In the options window that pops up, select the path curve and adjust settings such as Offset, Start Frame, End Frame, etc. Then it’s time to adjust the animation parameters and add additional animation effects, which will be changed a little further when you get down to the nitty-gritty.
