{"id":13440807,"url":"https://github.com/unmannedlab/LiDARNet","last_synced_at":"2025-03-20T10:32:47.463Z","repository":{"id":44481963,"uuid":"312344094","full_name":"unmannedlab/LiDARNet","owner":"unmannedlab","description":"LiDARNet: A Boundary-Aware Domain Adaptation Model for Point Cloud Semantic Segmentation","archived":false,"fork":false,"pushed_at":"2021-05-13T01:53:38.000Z","size":23429,"stargazers_count":32,"open_issues_count":3,"forks_count":3,"subscribers_count":6,"default_branch":"main","last_synced_at":"2024-08-01T03:32:02.967Z","etag":null,"topics":["3d-segmentation","domain-adaptation","lidar-point-cloud","semantic-segmentation"],"latest_commit_sha":null,"homepage":"https://unmannedlab.github.io/research/LiDARNet","language":null,"has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/unmannedlab.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null}},"created_at":"2020-11-12T17:10:31.000Z","updated_at":"2024-05-07T03:05:03.000Z","dependencies_parsed_at":"2022-08-26T00:23:53.286Z","dependency_job_id":null,"html_url":"https://github.com/unmannedlab/LiDARNet","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/unmannedlab%2FLiDARNet","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/unmannedlab%2FLiDARNet/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/unmannedlab%2FLiDARNet/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/unmannedlab%2FLiDARNet/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/unmannedlab","download_url":"https://codeload.github.com/unmannedlab/LiDARNet/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":244595480,"owners_count":20478480,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["3d-segmentation","domain-adaptation","lidar-point-cloud","semantic-segmentation"],"created_at":"2024-07-31T03:01:26.478Z","updated_at":"2025-03-20T10:32:47.449Z","avatar_url":"https://github.com/unmannedlab.png","language":null,"funding_links":[],"categories":["Others"],"sub_categories":[],"readme":"# LiDARNet: A Boundary-Aware Domain Adaptation Model for Point Cloud Semantic Segmentation\n\nWe present a boundary-aware domain adaptation model for LiDAR scan full-scene semantic segmentation (LiDARNet). Our model can extract both the domain private features and the domain shared features with a two branch structure.  We embedded Gated-SCNN into the segmentor component of LiDARNet to learn boundary information while learning to predict full-scene semantic segmentation labels. Moreover, we further reduce the domain gap by inducing the model to learn a mapping between two domains using the domain shared and private features. Besides, we introduce a new dataset ([SemanticUSL](https://unmannedlab.github.io/research/SemanticUSL)). The dataset has the same data format and ontology as SemanticKITTI. We conducted experiments on real-world datasets [SemanticKITTI](http://semantic-kitti.org/), [SemanticPOSS](poss.pku.edu.cn/semanticposs.html), and SemanticUSL, which have differences in channel distributions, reflectivity distributions, diversity of scenes, and sensors setup. Using our approach, we can get a single projection-based LiDAR full-scene semantic segmentation model working on both domains. Our model can keep almost the same performance on the source domain after adaptation and get an 8%-22% mIoU performance increase in the target domain.\n\n**Updates:**\n\n**The paper is released on [arXiv](https://arxiv.org/abs/2003.01174)**\n\n**The Code will come soon**\n\n## Results\n\n**Experiment I: From SemanticKITTI to SemanticPOSS and SemanticUSL** \n\n[![From SemanticKITTI to SemanticPOSS and SemanticUSL](https://img.youtube.com/vi/62C9cKzw3eY/0.jpg)](https://www.youtube.com/embed/62C9cKzw3eY)\n\n![LiDARNetkitti](images/LiDARNetkitti.png)\n\n\n\u003cbr/\u003e\u003cbr/\u003e\n\n**Experiment II: From SemanticPOSS to SemanticKITTI and SemanticUSL**\n\n[![From SemanticPOSS to SemanticKITTI and SemanticUSL](https://img.youtube.com/vi/jd-OaQ3jD5k/0.jpg)](https://www.youtube.com/embed/jd-OaQ3jD5k)\n\n![LiDARNetposs](images/LiDARNetposs.png)\n\n\n\u003cbr/\u003e\u003cbr/\u003e\n**Experiment III: From SemanticUSL to SemanticPOSS and SemanticKTTI**\n\n[![From SemanticPOSS to SemanticKITTI and SemanticUSL](https://img.youtube.com/vi/eRk7VJbQsRM/0.jpg)](https://www.youtube.com/embed/eRk7VJbQsRM)\n\n![LiDARNetusl](images/LiDARNetusl.png)\n\n## Citation\n```\n@misc{jiang2021lidarnet,\n      title={LiDARNet: A Boundary-Aware Domain Adaptation Model for Point Cloud Semantic Segmentation}, \n      author={Peng Jiang and Srikanth Saripalli},\n      year={2021},\n      eprint={2003.01174},\n      archivePrefix={arXiv},\n      primaryClass={cs.CV}\n}\n```\n## Related Work\n\n[SemanticUSL: A Dataset for Semantic Segmentation Domain Adatpation](https://unmannedlab.github.io/research/SemanticUSL)\n\n[RELLIS-3D: A Multi-modal Dataset for Off-Road Robotics](https://unmannedlab.github.io/research/RELLS-3D)\n\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Funmannedlab%2FLiDARNet","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Funmannedlab%2FLiDARNet","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Funmannedlab%2FLiDARNet/lists"}