AAAI2023,implementation of "READ: Large-Scale Neural Scene Rendering for Autonomous Driving", the experimental results are significantly better than Nerf-based methods
I try to obtain point cloud from kitti sequence 0, however, the generated pointcloud is distorted. The first diagram below is the point cloud generated by metashape, and the second is the point cloud provided by the author.
Here is the setting of metashape Align photo, so, what's wrong with the setting?
I try to obtain point cloud from kitti sequence 0, however, the generated pointcloud is distorted. The first diagram below is the point cloud generated by metashape, and the second is the point cloud provided by the author.![image](https://github.com/JOP-Lee/READ/assets/28439907/9846706f-617d-4edd-af75-1e5c7a772392)
Here is the setting of metashape Align photo, so, what's wrong with the setting?