MapCleaner: Efficiently Removing Moving Objects from Point Cloud Maps in Autonomous Driving Scenarios
Three-dimensional (3D) point cloud maps are widely used in autonomous driving scenarios. These maps are usually generated by accumulating sequential LiDAR scans. When generating a map, moving objects (such as vehicles or moving pedestrians) will leave long trails on the assembled map. This is undesi...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-09-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/14/18/4496 |
| _version_ | 1797482872895963136 |
|---|---|
| author | Hao Fu Hanzhang Xue Guanglei Xie |
| author_facet | Hao Fu Hanzhang Xue Guanglei Xie |
| author_sort | Hao Fu |
| collection | DOAJ |
| description | Three-dimensional (3D) point cloud maps are widely used in autonomous driving scenarios. These maps are usually generated by accumulating sequential LiDAR scans. When generating a map, moving objects (such as vehicles or moving pedestrians) will leave long trails on the assembled map. This is undesirable and reduces the map quality. In this paper, we propose MapCleaner, an approach that can effectively remove the moving objects from the map. MapCleaner first estimates a dense and continuous terrain surface, based on which the map point cloud is then divided into a noisy part below the terrain, the terrain, and the object part above the terrain. Next, a specifically designed moving points identification algorithm is performed on the object part to find moving objects. Experiments are performed on the SemanticKITTI dataset. Results show that the proposed MapCleaner outperforms state-of-the-art approaches on all five tested SemanticKITTI sequences. MapCleaner is a learning-free method and has few parameters to tune. It is also successfully evaluated on our own dataset collected with a different type of LiDAR. |
| first_indexed | 2024-03-09T22:38:46Z |
| format | Article |
| id | doaj.art-7fc63ec685cd4b86a466b9bc69b11f26 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-09T22:38:46Z |
| publishDate | 2022-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-7fc63ec685cd4b86a466b9bc69b11f262023-11-23T18:43:34ZengMDPI AGRemote Sensing2072-42922022-09-011418449610.3390/rs14184496MapCleaner: Efficiently Removing Moving Objects from Point Cloud Maps in Autonomous Driving ScenariosHao Fu0Hanzhang Xue1Guanglei Xie2College of Intelligence Science and Technology, National University of Defense Technology, Changsha 410073, ChinaCollege of Intelligence Science and Technology, National University of Defense Technology, Changsha 410073, ChinaCollege of Intelligence Science and Technology, National University of Defense Technology, Changsha 410073, ChinaThree-dimensional (3D) point cloud maps are widely used in autonomous driving scenarios. These maps are usually generated by accumulating sequential LiDAR scans. When generating a map, moving objects (such as vehicles or moving pedestrians) will leave long trails on the assembled map. This is undesirable and reduces the map quality. In this paper, we propose MapCleaner, an approach that can effectively remove the moving objects from the map. MapCleaner first estimates a dense and continuous terrain surface, based on which the map point cloud is then divided into a noisy part below the terrain, the terrain, and the object part above the terrain. Next, a specifically designed moving points identification algorithm is performed on the object part to find moving objects. Experiments are performed on the SemanticKITTI dataset. Results show that the proposed MapCleaner outperforms state-of-the-art approaches on all five tested SemanticKITTI sequences. MapCleaner is a learning-free method and has few parameters to tune. It is also successfully evaluated on our own dataset collected with a different type of LiDAR.https://www.mdpi.com/2072-4292/14/18/4496LiDAR point cloudmap cleaningautonomous drivingdynamic object |
| spellingShingle | Hao Fu Hanzhang Xue Guanglei Xie MapCleaner: Efficiently Removing Moving Objects from Point Cloud Maps in Autonomous Driving Scenarios Remote Sensing LiDAR point cloud map cleaning autonomous driving dynamic object |
| title | MapCleaner: Efficiently Removing Moving Objects from Point Cloud Maps in Autonomous Driving Scenarios |
| title_full | MapCleaner: Efficiently Removing Moving Objects from Point Cloud Maps in Autonomous Driving Scenarios |
| title_fullStr | MapCleaner: Efficiently Removing Moving Objects from Point Cloud Maps in Autonomous Driving Scenarios |
| title_full_unstemmed | MapCleaner: Efficiently Removing Moving Objects from Point Cloud Maps in Autonomous Driving Scenarios |
| title_short | MapCleaner: Efficiently Removing Moving Objects from Point Cloud Maps in Autonomous Driving Scenarios |
| title_sort | mapcleaner efficiently removing moving objects from point cloud maps in autonomous driving scenarios |
| topic | LiDAR point cloud map cleaning autonomous driving dynamic object |
| url | https://www.mdpi.com/2072-4292/14/18/4496 |
| work_keys_str_mv | AT haofu mapcleanerefficientlyremovingmovingobjectsfrompointcloudmapsinautonomousdrivingscenarios AT hanzhangxue mapcleanerefficientlyremovingmovingobjectsfrompointcloudmapsinautonomousdrivingscenarios AT guangleixie mapcleanerefficientlyremovingmovingobjectsfrompointcloudmapsinautonomousdrivingscenarios |