Thank you for your excellent work. However, when I ran the kitti dataset under the run_kitti_gs.yaml parameter you provided, I found that the SLAM localization(ATE...) accuracy was poor. May I ask what the reason is? Compared with the outstanding performance of PIN-SLAM, why does PINGS perform much worse on the kitti dataset?
Looking forward to your reply!
Thank you for your excellent work. However, when I ran the kitti dataset under the run_kitti_gs.yaml parameter you provided, I found that the SLAM localization(ATE...) accuracy was poor. May I ask what the reason is? Compared with the outstanding performance of PIN-SLAM, why does PINGS perform much worse on the kitti dataset?
Looking forward to your reply!